Model profile · Zhipu

glm-5.3-flash

Closed weightsmid · registry tag 2026 benchmark-derived

Thin verified coverage

Reads as thin verified coverage across the resolved source data. Capability fit excludes speed and cost.

Visible coverage: 44.5%
Verified coverage: 44.5%
Spread: 94.9%
Last verified: Oct 3, 2026

71%capability fit

textcodevisiondocument3 aliases43 official source links

Open compare

Data version

Current snapshot.

Published Oct 3, 2026Model list checked4 stale sources105 claim warnings9 providers · 1334 tracked models

The registry snapshot and page stamp are shown so a stale deploy is visible at a glance. View live data status.

Source-linked scores by benchmark

Each row keeps the benchmark source, source type, raw metric, and percentile inside its fair comparison set. Capability summaries exclude archived categories, speed, and cost; those rows remain below.

Thin verified coverageThis model currently reads as thin verified coverage across the resolved source data.

Chat / text35 benchmarks70.2%

Intelligence Index

AA · Chat / text · Combined

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #45 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 42
Percentile: 92.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.

92.1% percentile inside its fair comparison set

42Raw benchmark value

AA-Omniscience accuracy

AA · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #140 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 27.5%
Percentile: 69.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.

69.3% percentile inside its fair comparison set

27.5%Raw benchmark value

AA-Omniscience non-hallucination

AA · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #24 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 72.4%
Percentile: 94.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.

94.9% percentile inside its fair comparison set

72.4%Raw benchmark value

Input price

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #79 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $0.2 /1M input tokens
Percentile: 78.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.

78.2% percentile inside its fair comparison set

$0.2 /1M input tokensRaw benchmark value

Output price

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #78 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $0.5 /1M output tokens
Percentile: 78.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.

78.2% percentile inside its fair comparison set

$0.5 /1M output tokensRaw benchmark value

Output Speed

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #209 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 54.1 tokens/s
Percentile: 20.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

20.6% percentile inside its fair comparison set

54.1 tokens/sRaw benchmark value

Time to first token

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #150 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 3.21s
Percentile: 43.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.

43.6% percentile inside its fair comparison set

3.21sRaw benchmark value

Time to first answer token

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #200 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 40.16s
Percentile: 24%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.

24% percentile inside its fair comparison set

40.16sRaw benchmark value

Text Arena

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #39 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,473
Percentile: 89.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: overall. Source rank: #41. Votes: 22971. Organization: zai. License: MIT.

89.9% percentile inside its fair comparison set

1,473Raw benchmark valueCI 1,468 - 1,478

Text Arena · Creative Writing

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #62 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,433
Percentile: 83.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: creative_writing. Source rank: #69. Votes: 4870. Organization: zai. License: MIT.

83.7% percentile inside its fair comparison set

1,433Raw benchmark valueCI 1,424 - 1,443

Text Arena · English

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #43 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,478
Percentile: 88.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: english. Source rank: #44. Votes: 8546. Organization: zai. License: MIT.

88.8% percentile inside its fair comparison set

1,478Raw benchmark valueCI 1,471 - 1,485

Text Arena · Exclude Ties

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #41 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,481
Percentile: 89.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: exclude_ties. Source rank: #43. Votes: 16879. Organization: zai. License: MIT.

89.4% percentile inside its fair comparison set

1,481Raw benchmark valueCI 1,474 - 1,488

Text Arena · Hard Prompts

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,500
Percentile: 91.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: hard_prompts. Source rank: #33. Votes: 14880. Organization: zai. License: MIT.

91.5% percentile inside its fair comparison set

1,500Raw benchmark valueCI 1,494 - 1,506

Text Arena · Hard Prompts English

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #38 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,499
Percentile: 90.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: hard_prompts_english. Source rank: #39. Votes: 5515. Organization: zai. License: MIT.

90.1% percentile inside its fair comparison set

1,499Raw benchmark valueCI 1,490 - 1,507

Text Arena · Instruction Following

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #34 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,472
Percentile: 91.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: instruction_following. Source rank: #35. Votes: 8082. Organization: zai. License: MIT.

91.2% percentile inside its fair comparison set

1,472Raw benchmark valueCI 1,464 - 1,479

Text Arena · Longer Query

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #44 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,481
Percentile: 87.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: longer_query. Source rank: #45. Votes: 10474. Organization: zai. License: MIT.

87.9% percentile inside its fair comparison set

1,481Raw benchmark valueCI 1,474 - 1,488

Text Arena · Multi Turn

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #53 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,471
Percentile: 86.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: multi_turn. Source rank: #59. Votes: 3522. Organization: zai. License: MIT.

86.1% percentile inside its fair comparison set

1,471Raw benchmark valueCI 1,461 - 1,482

Text Arena · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #29 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,470
Percentile: 92.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: overall. Source rank: #30. Votes: 22971. Organization: zai. License: MIT.

92.6% percentile inside its fair comparison set

1,470Raw benchmark valueCI 1,464 - 1,475

Text Arena · Creative Writing · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #42 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,441
Percentile: 89%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: creative_writing. Source rank: #44. Votes: 4870. Organization: zai. License: MIT.

89% percentile inside its fair comparison set

1,441Raw benchmark valueCI 1,431 - 1,450

Text Arena · English · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,473
Percentile: 91.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: english. Source rank: #34. Votes: 8546. Organization: zai. License: MIT.

91.5% percentile inside its fair comparison set

1,473Raw benchmark valueCI 1,466 - 1,480

Text Arena · Exclude Ties · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #29 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,475
Percentile: 92.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: exclude_ties. Source rank: #30. Votes: 16879. Organization: zai. License: MIT.

92.6% percentile inside its fair comparison set

1,475Raw benchmark valueCI 1,468 - 1,482

Text Arena · Hard Prompts · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,491
Percentile: 95.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: hard_prompts. Source rank: #18. Votes: 14880. Organization: zai. License: MIT.

95.5% percentile inside its fair comparison set

1,491Raw benchmark valueCI 1,484 - 1,497

Text Arena · Hard Prompts English · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #23 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,488
Percentile: 94.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: hard_prompts_english. Source rank: #23. Votes: 5515. Organization: zai. License: MIT.

94.1% percentile inside its fair comparison set

1,488Raw benchmark valueCI 1,480 - 1,497

Text Arena · Instruction Following · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #20 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,478
Percentile: 94.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: instruction_following. Source rank: #20. Votes: 8082. Organization: zai. License: MIT.

94.9% percentile inside its fair comparison set

1,478Raw benchmark valueCI 1,470 - 1,485

Text Arena · Longer Query · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #22 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,483
Percentile: 94.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: longer_query. Source rank: #22. Votes: 10474. Organization: zai. License: MIT.

94.1% percentile inside its fair comparison set

1,483Raw benchmark valueCI 1,475 - 1,490

Text Arena · Multi Turn · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #36 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,467
Percentile: 90.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: multi_turn. Source rank: #38. Votes: 3522. Organization: zai. License: MIT.

90.6% percentile inside its fair comparison set

1,467Raw benchmark valueCI 1,457 - 1,477

Instruction following

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #66 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 52.8%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: IF. Tasks scored: 4.

0% percentile inside its fair comparison set

52.8%Raw benchmark value

Language

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #43 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 77.3%
Percentile: 35.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Language. Tasks scored: 3.

35.4% percentile inside its fair comparison set

77.3%Raw benchmark value

Paraphrase

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #65 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 48%
Percentile: 1.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: paraphrase. Category: IF.

1.5% percentile inside its fair comparison set

48%Raw benchmark value

Simplify

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #66 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 47%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: simplify. Category: IF.

0% percentile inside its fair comparison set

47%Raw benchmark value

Story generation

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #65 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 59.9%
Percentile: 1.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: story_generation. Category: IF.

1.5% percentile inside its fair comparison set

59.9%Raw benchmark value

Summarize

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #63 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 56.5%
Percentile: 4.6%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: summarize. Category: IF.

4.6% percentile inside its fair comparison set

56.5%Raw benchmark value

Connections

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #52 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 93.7%
Percentile: 21.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: connections. Category: Language.

21.5% percentile inside its fair comparison set

93.7%Raw benchmark value

Plot unscrambling

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #36 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 60.3%
Percentile: 46.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: plot_unscrambling. Category: Language.

46.2% percentile inside its fair comparison set

60.3%Raw benchmark value

Typos

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #45 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 78%
Percentile: 38.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: typos. Category: Language.

38.5% percentile inside its fair comparison set

78%Raw benchmark value

Coding21 benchmarks66.1%

SciCode

AA · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #71 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 51.6%
Percentile: 64.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `scicode`.

64.6% percentile inside its fair comparison set

51.6%Raw benchmark value

Text Arena · Coding

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #29 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,523
Percentile: 92.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: coding. Source rank: #29. Votes: 6157. Organization: zai. License: MIT.

92.5% percentile inside its fair comparison set

1,523Raw benchmark valueCI 1,515 - 1,532

Text Arena · Coding · No Style Control

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,506
Percentile: 95.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: coding. Source rank: #18. Votes: 6157. Organization: zai. License: MIT.

95.4% percentile inside its fair comparison set

1,506Raw benchmark valueCI 1,498 - 1,515

Vibe Code Bench v1.1

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #58 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 30.8%
Percentile: 38.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: vibe-code; provider: Zhipu AI.

38.7% percentile inside its fair comparison set

30.8%Raw benchmark valueCI 20.4% - 41.1%

WebDev Arena

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #24 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,616
Percentile: 81.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: overall. Source rank: #25. Votes: 10355. Organization: zai. License: MIT.

81.9% percentile inside its fair comparison set

1,616Raw benchmark valueCI 1,608 - 1,623

Code Arena · Webdev Html

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #26 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,578
Percentile: 80%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: webdev-html. Source rank: #27. Votes: 1562. Organization: zai. License: MIT.

80% percentile inside its fair comparison set

1,578Raw benchmark valueCI 1,561 - 1,595

Code Arena · Webdev React

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,635
Percentile: 85%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: webdev-react. Source rank: #19. Votes: 6797. Organization: zai. License: MIT.

85% percentile inside its fair comparison set

1,635Raw benchmark valueCI 1,627 - 1,644

Code Migration

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #45 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 20.5%
Percentile: 35.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: code-migration; provider: Zhipu AI.

35.3% percentile inside its fair comparison set

20.5%Raw benchmark valueCI 12.3% - 28.8%

IOI

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #21 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 52.5%
Percentile: 45.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: ioi; provider: Zhipu AI.

45.9% percentile inside its fair comparison set

52.5%Raw benchmark valueCI 47.4% - 57.6%

ProgramBench

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #37 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 0%
Percentile: 51.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: programbench; provider: Zhipu AI.

51.7% percentile inside its fair comparison set

0%Raw benchmark valueCI 0% - 0%

Terminal-Bench 2.1

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #36 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 62.9%
Percentile: 47.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: terminal-bench-2-1; provider: Fireworks AI.

47.8% percentile inside its fair comparison set

62.9%Raw benchmark valueCI 58.5% - 67.3%

Code Arena · Image To Webdev

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #9 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,588
Percentile: 81.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: image_to_webdev. Source rank: #10. Votes: 1600. Organization: zai. License: MIT.

81.8% percentile inside its fair comparison set

1,588Raw benchmark valueCI 1,572 - 1,603

SWE-bench Verified

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #10 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 92%
Percentile: 89%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: swebench; provider: Zhipu AI.

89% percentile inside its fair comparison set

92%Raw benchmark valueCI 89.6% - 94.4%

LiveCodeBench

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #66 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 80.5%
Percentile: 48%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: lcb; provider: Zhipu AI.

48% percentile inside its fair comparison set

80.5%Raw benchmark valueCI 78.5% - 82.6%

Agentic coding

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #23 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 56.8%
Percentile: 66.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Agentic Coding. Tasks scored: 3.

66.2% percentile inside its fair comparison set

56.8%Raw benchmark value

Coding

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #27 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 79%
Percentile: 61.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Coding. Tasks scored: 2.

61.5% percentile inside its fair comparison set

79%Raw benchmark value

JavaScript

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #41 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 63.6%
Percentile: 58.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: javascript. Category: Agentic Coding.

58.5% percentile inside its fair comparison set

63.6%Raw benchmark value

TypeScript

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #22 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 46.7%
Percentile: 72.3%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: typescript. Category: Agentic Coding.

72.3% percentile inside its fair comparison set

46.7%Raw benchmark value

Python

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #22 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 60%
Percentile: 80%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: python. Category: Agentic Coding.

80% percentile inside its fair comparison set

60%Raw benchmark value

Coding generation

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #36 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 77.5%
Percentile: 52.3%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: code_generation. Category: Coding.

52.3% percentile inside its fair comparison set

77.5%Raw benchmark value

Coding completion

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #24 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 80.4%
Percentile: 78.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: code_completion. Category: Coding.

78.5% percentile inside its fair comparison set

80.4%Raw benchmark value

Reasoning / math / science18 benchmarks53.1%

Humanity's Last Exam

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #63 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 39.9%
Percentile: 88.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `hle`.

88.3% percentile inside its fair comparison set

39.9%Raw benchmark value

GPQA

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #37 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 91.2%
Percentile: 92.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `gpqa`.

92.7% percentile inside its fair comparison set

91.2%Raw benchmark value

CritPt

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #62 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 15.4%
Percentile: 86.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `critpt`.

86.7% percentile inside its fair comparison set

15.4%Raw benchmark value

Text Arena · Math

AR · Reasoning / math / science · Human

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #13 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,503
Percentile: 96.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: math. Source rank: #13. Votes: 1131. Organization: zai. License: MIT.

96.7% percentile inside its fair comparison set

1,503Raw benchmark valueCI 1,485 - 1,521

Text Arena · Math · No Style Control

AR · Reasoning / math / science · Human

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #14 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,501
Percentile: 96.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: math. Source rank: #14. Votes: 1131. Organization: zai. License: MIT.

96.4% percentile inside its fair comparison set

1,501Raw benchmark valueCI 1,483 - 1,519

ProofBench

VALS-AI · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #35 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 21%
Percentile: 19%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: proof_bench; provider: Zhipu AI.

19% percentile inside its fair comparison set

21%Raw benchmark valueCI 13% - 29%

GPQA Diamond

VALS-AI · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #43 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 86.4%
Percentile: 66.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: gpqa; provider: Zhipu AI.

66.1% percentile inside its fair comparison set

86.4%Raw benchmark valueCI 82.2% - 90.5%

MMLU Pro

VALS-AI · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #49 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 86.1%
Percentile: 61.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: mmlu_pro; provider: Zhipu AI.

61.6% percentile inside its fair comparison set

86.1%Raw benchmark valueCI 85.4% - 86.7%

Mathematics

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #58 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 81.2%
Percentile: 12.3%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Mathematics. Tasks scored: 4.

12.3% percentile inside its fair comparison set

81.2%Raw benchmark value

Reasoning

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #56 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 77.6%
Percentile: 15.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Reasoning. Tasks scored: 4.

15.4% percentile inside its fair comparison set

77.6%Raw benchmark value

AMPS Hard

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #56 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 97%
Percentile: 32.3%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: AMPS_Hard. Category: Mathematics.

32.3% percentile inside its fair comparison set

97%Raw benchmark value

Integrals with game

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #61 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 45%
Percentile: 7.7%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: integrals_with_game. Category: Mathematics.

7.7% percentile inside its fair comparison set

45%Raw benchmark value

Math competition

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #61 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 93.1%
Percentile: 9.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: math_comp. Category: Mathematics.

9.2% percentile inside its fair comparison set

93.1%Raw benchmark value

Olympiad

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #40 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 89.8%
Percentile: 40%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: olympiad. Category: Mathematics.

40% percentile inside its fair comparison set

89.8%Raw benchmark value

Theory of mind

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #54 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 73.1%
Percentile: 23.1%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: theory_of_mind. Category: Reasoning.

23.1% percentile inside its fair comparison set

73.1%Raw benchmark value

Zebra puzzle

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #51 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 79.5%
Percentile: 24.6%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: zebra_puzzle. Category: Reasoning.

24.6% percentile inside its fair comparison set

79.5%Raw benchmark value

Spatial

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #52 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 96%
Percentile: 41.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: spatial. Category: Reasoning.

41.5% percentile inside its fair comparison set

96%Raw benchmark value

Logic with navigation

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #63 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 62%
Percentile: 6.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: logic_with_navigation. Category: Reasoning.

6.2% percentile inside its fair comparison set

62%Raw benchmark value

Professional reasoning30 benchmarks55.8%

Text Arena · Expert

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #28 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,511
Percentile: 91.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: expert. Source rank: #29. Votes: 2442. Organization: zai. License: MIT.

91.7% percentile inside its fair comparison set

1,511Raw benchmark valueCI 1,498 - 1,523

Text Arena · Industry Business And Management And Financial Operations

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #44 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,467
Percentile: 88.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_business_and_management_and_financial_operations. Source rank: #46. Votes: 4388. Organization: zai. License: MIT.

88.3% percentile inside its fair comparison set

1,467Raw benchmark valueCI 1,457 - 1,476

Text Arena · Industry Entertainment And Sports And Media

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #50 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,439
Percentile: 86.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_entertainment_and_sports_and_media. Source rank: #54. Votes: 5762. Organization: zai. License: MIT.

86.9% percentile inside its fair comparison set

1,439Raw benchmark valueCI 1,430 - 1,447

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #51 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,470
Percentile: 85.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_legal_and_government. Source rank: #56. Votes: 1927. Organization: zai. License: MIT.

85.7% percentile inside its fair comparison set

1,470Raw benchmark valueCI 1,456 - 1,484

Text Arena · Industry Life And Physical And Social Science

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,492
Percentile: 91.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_life_and_physical_and_social_science. Source rank: #34. Votes: 3729. Organization: zai. License: MIT.

91.4% percentile inside its fair comparison set

1,492Raw benchmark valueCI 1,482 - 1,502

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #24 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,495
Percentile: 93.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_mathematical. Source rank: #25. Votes: 1301. Organization: zai. License: MIT.

93.5% percentile inside its fair comparison set

1,495Raw benchmark valueCI 1,478 - 1,511

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #46 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,485
Percentile: 87%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_medicine_and_healthcare. Source rank: #47. Votes: 1500. Organization: zai. License: MIT.

87% percentile inside its fair comparison set

1,485Raw benchmark valueCI 1,469 - 1,500

Text Arena · Industry Software And It Services

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #26 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,514
Percentile: 93.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_software_and_it_services. Source rank: #26. Votes: 8872. Organization: zai. License: MIT.

93.4% percentile inside its fair comparison set

1,514Raw benchmark valueCI 1,507 - 1,521

Text Arena · Industry Writing And Literature And Language

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #53 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,448
Percentile: 86.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_writing_and_literature_and_language. Source rank: #58. Votes: 6316. Organization: zai. License: MIT.

86.1% percentile inside its fair comparison set

1,448Raw benchmark valueCI 1,440 - 1,456

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #19 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,512
Percentile: 94.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: expert. Source rank: #19. Votes: 2442. Organization: zai. License: MIT.

94.5% percentile inside its fair comparison set

1,512Raw benchmark valueCI 1,500 - 1,525

Text Arena · Industry Business And Management And Financial Operations · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,458
Percentile: 91.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_business_and_management_and_financial_operations. Source rank: #34. Votes: 4388. Organization: zai. License: MIT.

91.3% percentile inside its fair comparison set

1,458Raw benchmark valueCI 1,449 - 1,468

Text Arena · Industry Entertainment And Sports And Media · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #32 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,442
Percentile: 91.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_entertainment_and_sports_and_media. Source rank: #34. Votes: 5762. Organization: zai. License: MIT.

91.7% percentile inside its fair comparison set

1,442Raw benchmark valueCI 1,433 - 1,451

Text Arena · Industry Legal And Government · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #35 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,473
Percentile: 90.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_legal_and_government. Source rank: #37. Votes: 1927. Organization: zai. License: MIT.

90.3% percentile inside its fair comparison set

1,473Raw benchmark valueCI 1,459 - 1,487

Text Arena · Industry Life And Physical And Social Science · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #25 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,487
Percentile: 93.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_life_and_physical_and_social_science. Source rank: #26. Votes: 3729. Organization: zai. License: MIT.

93.6% percentile inside its fair comparison set

1,487Raw benchmark valueCI 1,477 - 1,496

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #20 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,496
Percentile: 94.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_mathematical. Source rank: #20. Votes: 1301. Organization: zai. License: MIT.

94.7% percentile inside its fair comparison set

1,496Raw benchmark valueCI 1,479 - 1,512

Text Arena · Industry Medicine And Healthcare · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,470
Percentile: 90.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_medicine_and_healthcare. Source rank: #34. Votes: 1500. Organization: zai. License: MIT.

90.8% percentile inside its fair comparison set

1,470Raw benchmark valueCI 1,455 - 1,486

Text Arena · Industry Software And It Services · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,500
Percentile: 95.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_software_and_it_services. Source rank: #18. Votes: 8872. Organization: zai. License: MIT.

95.5% percentile inside its fair comparison set

1,500Raw benchmark valueCI 1,493 - 1,507

Text Arena · Industry Writing And Literature And Language · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #37 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,452
Percentile: 90.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_writing_and_literature_and_language. Source rank: #39. Votes: 6316. Organization: zai. License: MIT.

90.4% percentile inside its fair comparison set

1,452Raw benchmark valueCI 1,444 - 1,460

Legal Research Bench

VALS-AI · Professional reasoning · Objective

Applied legal research tasks.

Rank #15 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 45.2%
Percentile: 79.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_research; provider: Zhipu AI.

79.4% percentile inside its fair comparison set

45.2%Raw benchmark valueCI 38.4% - 52%

Harvey's Legal Agent Benchmark

VALS-AI · Professional reasoning · Objective

Completing legal work with documents, spreadsheets, presentations, and file-system tools.

Rank #25 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 6.7%
Percentile: 68.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Zhipu AI.

68.1% percentile inside its fair comparison set

6.7%Raw benchmark valueCI 3.4% - 9.9%

LegalBench

VALS-AI · Professional reasoning · Objective

Academic legal reasoning tasks.

Rank #44 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 83.9%
Percentile: 68.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

68.1% percentile inside its fair comparison set

83.9%Raw benchmark valueCI 83.1% - 84.8%

Finance Agent v2

VALS-AI · Professional reasoning · Objective

Core financial analyst tasks for agentic models.

Rank #11 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 57.9%
Percentile: 85.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: fabv2; provider: Fireworks AI.

85.5% percentile inside its fair comparison set

57.9%Raw benchmark valueCI 53.8% - 61.9%

MedScribe

VALS-AI · Professional reasoning · Objective

Administrative documentation support for doctors.

Rank #7 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 88.9%
Percentile: 93.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: medscribe; provider: Zhipu AI.

93.8% percentile inside its fair comparison set

88.9%Raw benchmark valueCI 85.2% - 92.7%

SkillsBench

VALS-AI · Professional reasoning · Objective

Applied professional skills tasks.

Rank #29 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 40.2%
Percentile: 15.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: skillsbench; provider: Fireworks AI.

15.2% percentile inside its fair comparison set

40.2%Raw benchmark valueCI 31.5% - 48.8%

TaxEval v2

VALS-AI · Professional reasoning · Objective

Answer quality on tax questions and responses.

Rank #16 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 75.6%
Percentile: 88.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: tax_eval_v2; provider: Zhipu AI.

88.4% percentile inside its fair comparison set

75.6%Raw benchmark valueCI 74% - 77.2%

Data analysis

LB · Professional reasoning · Objective

Structured data manipulation and table reasoning accuracy.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 76.4%
Percentile: 50.8%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Data Analysis. Tasks scored: 3.

50.8% percentile inside its fair comparison set

76.4%Raw benchmark value

Overall

LB · Professional reasoning · Objective

Average objective performance across LiveBench's current public category mix.

Rank #52 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 71.6%
Percentile: 21.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category averages included: 7.

21.5% percentile inside its fair comparison set

71.6%Raw benchmark value

Consecutive events

LB · Professional reasoning · Objective

Objective consecutive events score in LiveBench.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 84.3%
Percentile: 50.8%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: consecutive_events. Category: Data Analysis.

50.8% percentile inside its fair comparison set

84.3%Raw benchmark value

Table join

LB · Professional reasoning · Objective

Objective table join score in LiveBench.

Rank #28 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 48.8%
Percentile: 58.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: tablejoin. Category: Data Analysis.

58.5% percentile inside its fair comparison set

48.8%Raw benchmark value

Table reformat

LB · Professional reasoning · Objective

Objective table reformat score in LiveBench.

Rank #59 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 96.1%
Percentile: 20%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: tablereformat. Category: Data Analysis.

20% percentile inside its fair comparison set

96.1%Raw benchmark value

Long context1 benchmark88.6%

Long Context Reasoning

AA · Long context · Objective

It checks whether long-context claims survive contact with retrieval, memory, or long-document tasks.

Rank #53 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 80%
Percentile: 88.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `lcr`.

88.6% percentile inside its fair comparison set

80%Raw benchmark value

Vision understanding15 benchmarks77.4%

Vision Arena

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #27 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,278
Percentile: 81.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: overall. Source rank: #30. Votes: 5479. Organization: zai. License: MIT.

81.3% percentile inside its fair comparison set

1,278Raw benchmark valueCI 1,269 - 1,288

Vision Arena · Creative Writing Vision

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #28 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,259
Percentile: 67.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: creative_writing_vision. Source rank: #32. Votes: 493. Organization: zai. License: MIT.

67.1% percentile inside its fair comparison set

1,259Raw benchmark valueCI 1,232 - 1,287

Vision Arena · Diagram

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,313
Percentile: 82.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: diagram. Source rank: #20. Votes: 1299. Organization: zai. License: MIT.

82.8% percentile inside its fair comparison set

1,313Raw benchmark valueCI 1,296 - 1,330

Vision Arena · English

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #30 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,272
Percentile: 79%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: english. Source rank: #33. Votes: 1731. Organization: zai. License: MIT.

79% percentile inside its fair comparison set

1,272Raw benchmark valueCI 1,257 - 1,287

Vision Arena · Homework

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #11 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,331
Percentile: 89.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: homework. Source rank: #12. Votes: 616. Organization: zai. License: MIT.

89.6% percentile inside its fair comparison set

1,331Raw benchmark valueCI 1,307 - 1,355

Vision Arena · Humor

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #31 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,251
Percentile: 58.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: humor. Source rank: #35. Votes: 242. Organization: zai. License: MIT.

58.9% percentile inside its fair comparison set

1,251Raw benchmark valueCI 1,213 - 1,290

Vision Arena · Ocr

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #28 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,291
Percentile: 73%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: ocr. Source rank: #30. Votes: 3682. Organization: zai. License: MIT.

73% percentile inside its fair comparison set

1,291Raw benchmark valueCI 1,280 - 1,301

Vision Arena · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #17 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,301
Percentile: 88.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: overall. Source rank: #18. Votes: 5479. Organization: zai. License: MIT.

88.5% percentile inside its fair comparison set

1,301Raw benchmark valueCI 1,292 - 1,310

Vision Arena · Creative Writing Vision · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #21 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,286
Percentile: 75.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: creative_writing_vision. Source rank: #23. Votes: 493. Organization: zai. License: MIT.

75.6% percentile inside its fair comparison set

1,286Raw benchmark valueCI 1,258 - 1,313

Vision Arena · Diagram · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #15 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,323
Percentile: 85.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: diagram. Source rank: #16. Votes: 1299. Organization: zai. License: MIT.

85.9% percentile inside its fair comparison set

1,323Raw benchmark valueCI 1,306 - 1,340

Vision Arena · English · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #15 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,297
Percentile: 89.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: english. Source rank: #16. Votes: 1731. Organization: zai. License: MIT.

89.9% percentile inside its fair comparison set

1,297Raw benchmark valueCI 1,282 - 1,312

Vision Arena · Homework · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #7 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,343
Percentile: 93.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: homework. Source rank: #7. Votes: 616. Organization: zai. License: MIT.

93.8% percentile inside its fair comparison set

1,343Raw benchmark valueCI 1,319 - 1,367

Vision Arena · Humor · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #25 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,274
Percentile: 67.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: humor. Source rank: #28. Votes: 242. Organization: zai. License: MIT.

67.1% percentile inside its fair comparison set

1,274Raw benchmark valueCI 1,236 - 1,313

Vision Arena · Ocr · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,308
Percentile: 83%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: ocr. Source rank: #19. Votes: 3682. Organization: zai. License: MIT.

83% percentile inside its fair comparison set

1,308Raw benchmark valueCI 1,297 - 1,318

MMMU Pro

VALS-AI · Vision understanding · Objective

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #23 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 86%
Percentile: 73.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: mmmu; provider: Zhipu AI.

73.5% percentile inside its fair comparison set

86%Raw benchmark valueCI 84.4% - 87.6%

Multilingual20 benchmarksn/a

Text Arena · Chinese

AR · Multilingual · Human

Observed user preference in Arena's Text Arena chinese leaderboard.

Rank #21 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,529
Percentile: 94.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: chinese. Source rank: #21. Votes: 1648. Organization: zai. License: MIT.

94.3% percentile inside its fair comparison set

1,529Raw benchmark valueCI 1,513 - 1,544

Text Arena · French

AR · Multilingual · Human

Observed user preference in Arena's Text Arena french leaderboard.

Rank #15 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,507
Percentile: 94.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: french. Source rank: #15. Votes: 765. Organization: zai. License: MIT.

94.6% percentile inside its fair comparison set

1,507Raw benchmark valueCI 1,485 - 1,530

Text Arena · German

AR · Multilingual · Human

Observed user preference in Arena's Text Arena german leaderboard.

Rank #44 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,465
Percentile: 84.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: german. Source rank: #48. Votes: 395. Organization: zai. License: MIT.

84.2% percentile inside its fair comparison set

1,465Raw benchmark valueCI 1,435 - 1,494

Text Arena · Japanese

AR · Multilingual · Human

Observed user preference in Arena's Text Arena japanese leaderboard.

Rank #45 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,433
Percentile: 81.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: japanese. Source rank: #50. Votes: 408. Organization: zai. License: MIT.

81.9% percentile inside its fair comparison set

1,433Raw benchmark valueCI 1,402 - 1,464

Text Arena · Korean

AR · Multilingual · Human

Observed user preference in Arena's Text Arena korean leaderboard.

Rank #30 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,438
Percentile: 88.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: korean. Source rank: #31. Votes: 397. Organization: zai. License: MIT.

88.1% percentile inside its fair comparison set

1,438Raw benchmark valueCI 1,408 - 1,469

Text Arena · Non English

AR · Multilingual · Human

Observed user preference in Arena's Text Arena non_english leaderboard.

Rank #35 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,463
Percentile: 91%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: non_english. Source rank: #37. Votes: 14422. Organization: zai. License: MIT.

91% percentile inside its fair comparison set

1,463Raw benchmark valueCI 1,457 - 1,469

Text Arena · Polish

AR · Multilingual · Human

Observed user preference in Arena's Text Arena polish leaderboard.

Rank #60 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,447
Percentile: 70.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: polish. Source rank: #73. Votes: 370. Organization: zai. License: MIT.

70.4% percentile inside its fair comparison set

1,447Raw benchmark valueCI 1,417 - 1,476

Text Arena · Russian

AR · Multilingual · Human

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #36 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,475
Percentile: 89.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: russian. Source rank: #40. Votes: 3160. Organization: zai. License: MIT.

89.7% percentile inside its fair comparison set

1,475Raw benchmark valueCI 1,464 - 1,486

Text Arena · Spanish

AR · Multilingual · Human

Observed user preference in Arena's Text Arena spanish leaderboard.

Rank #37 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,466
Percentile: 86.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: spanish. Source rank: #39. Votes: 792. Organization: zai. License: MIT.

86.1% percentile inside its fair comparison set

1,466Raw benchmark valueCI 1,444 - 1,487

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena chinese leaderboard.

Rank #23 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,528
Percentile: 93.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: chinese. Source rank: #23. Votes: 1648. Organization: zai. License: MIT.

93.8% percentile inside its fair comparison set

1,528Raw benchmark valueCI 1,513 - 1,543

Text Arena · French · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena french leaderboard.

Rank #13 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,498
Percentile: 95.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: french. Source rank: #13. Votes: 765. Organization: zai. License: MIT.

95.3% percentile inside its fair comparison set

1,498Raw benchmark valueCI 1,476 - 1,521

Text Arena · German · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena german leaderboard.

Rank #37 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,458
Percentile: 86.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: german. Source rank: #40. Votes: 395. Organization: zai. License: MIT.

86.8% percentile inside its fair comparison set

1,458Raw benchmark valueCI 1,428 - 1,488

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena japanese leaderboard.

Rank #40 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,425
Percentile: 84%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: japanese. Source rank: #43. Votes: 408. Organization: zai. License: MIT.

84% percentile inside its fair comparison set

1,425Raw benchmark valueCI 1,394 - 1,456

Text Arena · Korean · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena korean leaderboard.

Rank #25 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,439
Percentile: 90.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: korean. Source rank: #27. Votes: 397. Organization: zai. License: MIT.

90.2% percentile inside its fair comparison set

1,439Raw benchmark valueCI 1,408 - 1,469

Text Arena · Non English · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena non_english leaderboard.

Rank #28 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,461
Percentile: 92.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: non_english. Source rank: #29. Votes: 14422. Organization: zai. License: MIT.

92.8% percentile inside its fair comparison set

1,461Raw benchmark valueCI 1,455 - 1,467

Text Arena · Polish · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena polish leaderboard.

Rank #55 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,439
Percentile: 72.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: polish. Source rank: #62. Votes: 370. Organization: zai. License: MIT.

72.9% percentile inside its fair comparison set

1,439Raw benchmark valueCI 1,409 - 1,469

Text Arena · Russian · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #29 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,471
Percentile: 91.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: russian. Source rank: #31. Votes: 3160. Organization: zai. License: MIT.

91.8% percentile inside its fair comparison set

1,471Raw benchmark valueCI 1,460 - 1,482

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena spanish leaderboard.

Rank #27 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,469
Percentile: 90%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: spanish. Source rank: #28. Votes: 792. Organization: zai. License: MIT.

90% percentile inside its fair comparison set

1,469Raw benchmark valueCI 1,448 - 1,490

Vision Arena · Chinese

AR · Multilingual · Human

Observed user preference in Arena's Vision Arena chinese leaderboard.

Rank #8 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,349
Percentile: 93.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: chinese. Source rank: #8. Votes: 231. Organization: zai. License: MIT.

93.3% percentile inside its fair comparison set

1,349Raw benchmark valueCI 1,311 - 1,387

Vision Arena · Chinese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Vision Arena chinese leaderboard.

Rank #5 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,373
Percentile: 96.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: chinese. Source rank: #6. Votes: 231. Organization: zai. License: MIT.

96.2% percentile inside its fair comparison set

1,373Raw benchmark valueCI 1,335 - 1,411

Source links and registry checks

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Artificial Analysis

Oct 3, 2026

source →

official

LiveBench

Oct 3, 2026

source →

Model profile · Zhipu

glm-5.3-flash

Closed weightsmid · registry tag 2026 benchmark-derived

Thin verified coverage

Reads as thin verified coverage across the resolved source data. Capability fit excludes speed and cost.

Visible coverage: 44.5%
Verified coverage: 44.5%
Spread: 94.9%
Last verified: Oct 3, 2026

71%capability fit

textcodevisiondocument3 aliases43 official source links

Open compare

Data version

Current snapshot.

Published Oct 3, 2026Model list checked4 stale sources105 claim warnings9 providers · 1334 tracked models

The registry snapshot and page stamp are shown so a stale deploy is visible at a glance. View live data status.

Source-linked scores by benchmark

Thin verified coverageThis model currently reads as thin verified coverage across the resolved source data.

Chat / text35 benchmarks70.2%

Intelligence Index

AA · Chat / text · Combined

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #45 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 42
Percentile: 92.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.

92.1% percentile inside its fair comparison set

42Raw benchmark value

AA-Omniscience accuracy

AA · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #140 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 27.5%
Percentile: 69.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.

69.3% percentile inside its fair comparison set

27.5%Raw benchmark value

AA-Omniscience non-hallucination

AA · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #24 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 72.4%
Percentile: 94.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.

94.9% percentile inside its fair comparison set

72.4%Raw benchmark value

Input price

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #79 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $0.2 /1M input tokens
Percentile: 78.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.

78.2% percentile inside its fair comparison set

$0.2 /1M input tokensRaw benchmark value

Output price

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #78 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $0.5 /1M output tokens
Percentile: 78.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.

78.2% percentile inside its fair comparison set

$0.5 /1M output tokensRaw benchmark value

Output Speed

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #209 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 54.1 tokens/s
Percentile: 20.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

20.6% percentile inside its fair comparison set

54.1 tokens/sRaw benchmark value

Time to first token

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #150 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 3.21s
Percentile: 43.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.

43.6% percentile inside its fair comparison set

3.21sRaw benchmark value

Time to first answer token

AA · Chat / text · Speed / cost

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #200 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 40.16s
Percentile: 24%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.

24% percentile inside its fair comparison set

40.16sRaw benchmark value

Text Arena

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #39 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,473
Percentile: 89.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: overall. Source rank: #41. Votes: 22971. Organization: zai. License: MIT.

89.9% percentile inside its fair comparison set

1,473Raw benchmark valueCI 1,468 - 1,478

Text Arena · Creative Writing

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #62 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,433
Percentile: 83.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: creative_writing. Source rank: #69. Votes: 4870. Organization: zai. License: MIT.

83.7% percentile inside its fair comparison set

1,433Raw benchmark valueCI 1,424 - 1,443

Text Arena · English

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #43 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,478
Percentile: 88.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: english. Source rank: #44. Votes: 8546. Organization: zai. License: MIT.

88.8% percentile inside its fair comparison set

1,478Raw benchmark valueCI 1,471 - 1,485

Text Arena · Exclude Ties

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #41 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,481
Percentile: 89.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: exclude_ties. Source rank: #43. Votes: 16879. Organization: zai. License: MIT.

89.4% percentile inside its fair comparison set

1,481Raw benchmark valueCI 1,474 - 1,488

Text Arena · Hard Prompts

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,500
Percentile: 91.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: hard_prompts. Source rank: #33. Votes: 14880. Organization: zai. License: MIT.

91.5% percentile inside its fair comparison set

1,500Raw benchmark valueCI 1,494 - 1,506

Text Arena · Hard Prompts English

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #38 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,499
Percentile: 90.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: hard_prompts_english. Source rank: #39. Votes: 5515. Organization: zai. License: MIT.

90.1% percentile inside its fair comparison set

1,499Raw benchmark valueCI 1,490 - 1,507

Text Arena · Instruction Following

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #34 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,472
Percentile: 91.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: instruction_following. Source rank: #35. Votes: 8082. Organization: zai. License: MIT.

91.2% percentile inside its fair comparison set

1,472Raw benchmark valueCI 1,464 - 1,479

Text Arena · Longer Query

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #44 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,481
Percentile: 87.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: longer_query. Source rank: #45. Votes: 10474. Organization: zai. License: MIT.

87.9% percentile inside its fair comparison set

1,481Raw benchmark valueCI 1,474 - 1,488

Text Arena · Multi Turn

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #53 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,471
Percentile: 86.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: multi_turn. Source rank: #59. Votes: 3522. Organization: zai. License: MIT.

86.1% percentile inside its fair comparison set

1,471Raw benchmark valueCI 1,461 - 1,482

Text Arena · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #29 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,470
Percentile: 92.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: overall. Source rank: #30. Votes: 22971. Organization: zai. License: MIT.

92.6% percentile inside its fair comparison set

1,470Raw benchmark valueCI 1,464 - 1,475

Text Arena · Creative Writing · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #42 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,441
Percentile: 89%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: creative_writing. Source rank: #44. Votes: 4870. Organization: zai. License: MIT.

89% percentile inside its fair comparison set

1,441Raw benchmark valueCI 1,431 - 1,450

Text Arena · English · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,473
Percentile: 91.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: english. Source rank: #34. Votes: 8546. Organization: zai. License: MIT.

91.5% percentile inside its fair comparison set

1,473Raw benchmark valueCI 1,466 - 1,480

Text Arena · Exclude Ties · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #29 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,475
Percentile: 92.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: exclude_ties. Source rank: #30. Votes: 16879. Organization: zai. License: MIT.

92.6% percentile inside its fair comparison set

1,475Raw benchmark valueCI 1,468 - 1,482

Text Arena · Hard Prompts · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,491
Percentile: 95.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: hard_prompts. Source rank: #18. Votes: 14880. Organization: zai. License: MIT.

95.5% percentile inside its fair comparison set

1,491Raw benchmark valueCI 1,484 - 1,497

Text Arena · Hard Prompts English · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #23 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,488
Percentile: 94.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: hard_prompts_english. Source rank: #23. Votes: 5515. Organization: zai. License: MIT.

94.1% percentile inside its fair comparison set

1,488Raw benchmark valueCI 1,480 - 1,497

Text Arena · Instruction Following · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #20 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,478
Percentile: 94.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: instruction_following. Source rank: #20. Votes: 8082. Organization: zai. License: MIT.

94.9% percentile inside its fair comparison set

1,478Raw benchmark valueCI 1,470 - 1,485

Text Arena · Longer Query · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #22 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,483
Percentile: 94.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: longer_query. Source rank: #22. Votes: 10474. Organization: zai. License: MIT.

94.1% percentile inside its fair comparison set

1,483Raw benchmark valueCI 1,475 - 1,490

Text Arena · Multi Turn · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #36 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,467
Percentile: 90.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: multi_turn. Source rank: #38. Votes: 3522. Organization: zai. License: MIT.

90.6% percentile inside its fair comparison set

1,467Raw benchmark valueCI 1,457 - 1,477

Instruction following

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #66 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 52.8%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: IF. Tasks scored: 4.

0% percentile inside its fair comparison set

52.8%Raw benchmark value

Language

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #43 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 77.3%
Percentile: 35.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Language. Tasks scored: 3.

35.4% percentile inside its fair comparison set

77.3%Raw benchmark value

Paraphrase

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #65 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 48%
Percentile: 1.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: paraphrase. Category: IF.

1.5% percentile inside its fair comparison set

48%Raw benchmark value

Simplify

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #66 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 47%
Percentile: 0%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: simplify. Category: IF.

0% percentile inside its fair comparison set

47%Raw benchmark value

Story generation

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #65 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 59.9%
Percentile: 1.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: story_generation. Category: IF.

1.5% percentile inside its fair comparison set

59.9%Raw benchmark value

Summarize

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #63 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 56.5%
Percentile: 4.6%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: summarize. Category: IF.

4.6% percentile inside its fair comparison set

56.5%Raw benchmark value

Connections

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #52 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 93.7%
Percentile: 21.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: connections. Category: Language.

21.5% percentile inside its fair comparison set

93.7%Raw benchmark value

Plot unscrambling

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #36 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 60.3%
Percentile: 46.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: plot_unscrambling. Category: Language.

46.2% percentile inside its fair comparison set

60.3%Raw benchmark value

Typos

LB · Chat / text · Objective

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #45 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 78%
Percentile: 38.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: typos. Category: Language.

38.5% percentile inside its fair comparison set

78%Raw benchmark value

Coding21 benchmarks66.1%

SciCode

AA · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #71 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 51.6%
Percentile: 64.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `scicode`.

64.6% percentile inside its fair comparison set

51.6%Raw benchmark value

Text Arena · Coding

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #29 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,523
Percentile: 92.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: coding. Source rank: #29. Votes: 6157. Organization: zai. License: MIT.

92.5% percentile inside its fair comparison set

1,523Raw benchmark valueCI 1,515 - 1,532

Text Arena · Coding · No Style Control

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,506
Percentile: 95.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: coding. Source rank: #18. Votes: 6157. Organization: zai. License: MIT.

95.4% percentile inside its fair comparison set

1,506Raw benchmark valueCI 1,498 - 1,515

Vibe Code Bench v1.1

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #58 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 30.8%
Percentile: 38.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: vibe-code; provider: Zhipu AI.

38.7% percentile inside its fair comparison set

30.8%Raw benchmark valueCI 20.4% - 41.1%

WebDev Arena

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #24 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,616
Percentile: 81.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: overall. Source rank: #25. Votes: 10355. Organization: zai. License: MIT.

81.9% percentile inside its fair comparison set

1,616Raw benchmark valueCI 1,608 - 1,623

Code Arena · Webdev Html

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #26 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,578
Percentile: 80%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: webdev-html. Source rank: #27. Votes: 1562. Organization: zai. License: MIT.

80% percentile inside its fair comparison set

1,578Raw benchmark valueCI 1,561 - 1,595

Code Arena · Webdev React

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,635
Percentile: 85%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: webdev-react. Source rank: #19. Votes: 6797. Organization: zai. License: MIT.

85% percentile inside its fair comparison set

1,635Raw benchmark valueCI 1,627 - 1,644

Code Migration

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #45 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 20.5%
Percentile: 35.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: code-migration; provider: Zhipu AI.

35.3% percentile inside its fair comparison set

20.5%Raw benchmark valueCI 12.3% - 28.8%

IOI

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #21 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 52.5%
Percentile: 45.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: ioi; provider: Zhipu AI.

45.9% percentile inside its fair comparison set

52.5%Raw benchmark valueCI 47.4% - 57.6%

ProgramBench

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #37 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 0%
Percentile: 51.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: programbench; provider: Zhipu AI.

51.7% percentile inside its fair comparison set

0%Raw benchmark valueCI 0% - 0%

Terminal-Bench 2.1

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #36 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 62.9%
Percentile: 47.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: terminal-bench-2-1; provider: Fireworks AI.

47.8% percentile inside its fair comparison set

62.9%Raw benchmark valueCI 58.5% - 67.3%

Code Arena · Image To Webdev

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #9 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,588
Percentile: 81.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: image_to_webdev. Source rank: #10. Votes: 1600. Organization: zai. License: MIT.

81.8% percentile inside its fair comparison set

1,588Raw benchmark valueCI 1,572 - 1,603

SWE-bench Verified

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #10 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 92%
Percentile: 89%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: swebench; provider: Zhipu AI.

89% percentile inside its fair comparison set

92%Raw benchmark valueCI 89.6% - 94.4%

LiveCodeBench

VALS-AI · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #66 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 80.5%
Percentile: 48%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: lcb; provider: Zhipu AI.

48% percentile inside its fair comparison set

80.5%Raw benchmark valueCI 78.5% - 82.6%

Agentic coding

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #23 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 56.8%
Percentile: 66.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Agentic Coding. Tasks scored: 3.

66.2% percentile inside its fair comparison set

56.8%Raw benchmark value

Coding

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #27 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 79%
Percentile: 61.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Coding. Tasks scored: 2.

61.5% percentile inside its fair comparison set

79%Raw benchmark value

JavaScript

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #41 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 63.6%
Percentile: 58.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: javascript. Category: Agentic Coding.

58.5% percentile inside its fair comparison set

63.6%Raw benchmark value

TypeScript

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #22 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 46.7%
Percentile: 72.3%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: typescript. Category: Agentic Coding.

72.3% percentile inside its fair comparison set

46.7%Raw benchmark value

Python

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #22 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 60%
Percentile: 80%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: python. Category: Agentic Coding.

80% percentile inside its fair comparison set

60%Raw benchmark value

Coding generation

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #36 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 77.5%
Percentile: 52.3%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: code_generation. Category: Coding.

52.3% percentile inside its fair comparison set

77.5%Raw benchmark value

Coding completion

LB · Coding · Objective

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #24 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 80.4%
Percentile: 78.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: code_completion. Category: Coding.

78.5% percentile inside its fair comparison set

80.4%Raw benchmark value

Reasoning / math / science18 benchmarks53.1%

Humanity's Last Exam

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #63 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 39.9%
Percentile: 88.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `hle`.

88.3% percentile inside its fair comparison set

39.9%Raw benchmark value

GPQA

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #37 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 91.2%
Percentile: 92.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `gpqa`.

92.7% percentile inside its fair comparison set

91.2%Raw benchmark value

CritPt

AA · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #62 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 15.4%
Percentile: 86.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `critpt`.

86.7% percentile inside its fair comparison set

15.4%Raw benchmark value

Text Arena · Math

AR · Reasoning / math / science · Human

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #13 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,503
Percentile: 96.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: math. Source rank: #13. Votes: 1131. Organization: zai. License: MIT.

96.7% percentile inside its fair comparison set

1,503Raw benchmark valueCI 1,485 - 1,521

Text Arena · Math · No Style Control

AR · Reasoning / math / science · Human

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #14 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,501
Percentile: 96.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: math. Source rank: #14. Votes: 1131. Organization: zai. License: MIT.

96.4% percentile inside its fair comparison set

1,501Raw benchmark valueCI 1,483 - 1,519

ProofBench

VALS-AI · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #35 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 21%
Percentile: 19%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: proof_bench; provider: Zhipu AI.

19% percentile inside its fair comparison set

21%Raw benchmark valueCI 13% - 29%

GPQA Diamond

VALS-AI · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #43 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 86.4%
Percentile: 66.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: gpqa; provider: Zhipu AI.

66.1% percentile inside its fair comparison set

86.4%Raw benchmark valueCI 82.2% - 90.5%

MMLU Pro

VALS-AI · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #49 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 86.1%
Percentile: 61.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: mmlu_pro; provider: Zhipu AI.

61.6% percentile inside its fair comparison set

86.1%Raw benchmark valueCI 85.4% - 86.7%

Mathematics

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #58 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 81.2%
Percentile: 12.3%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Mathematics. Tasks scored: 4.

12.3% percentile inside its fair comparison set

81.2%Raw benchmark value

Reasoning

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #56 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 77.6%
Percentile: 15.4%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Reasoning. Tasks scored: 4.

15.4% percentile inside its fair comparison set

77.6%Raw benchmark value

AMPS Hard

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #56 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 97%
Percentile: 32.3%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: AMPS_Hard. Category: Mathematics.

32.3% percentile inside its fair comparison set

97%Raw benchmark value

Integrals with game

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #61 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 45%
Percentile: 7.7%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: integrals_with_game. Category: Mathematics.

7.7% percentile inside its fair comparison set

45%Raw benchmark value

Math competition

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #61 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 93.1%
Percentile: 9.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: math_comp. Category: Mathematics.

9.2% percentile inside its fair comparison set

93.1%Raw benchmark value

Olympiad

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #40 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 89.8%
Percentile: 40%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: olympiad. Category: Mathematics.

40% percentile inside its fair comparison set

89.8%Raw benchmark value

Theory of mind

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #54 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 73.1%
Percentile: 23.1%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: theory_of_mind. Category: Reasoning.

23.1% percentile inside its fair comparison set

73.1%Raw benchmark value

Zebra puzzle

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #51 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 79.5%
Percentile: 24.6%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: zebra_puzzle. Category: Reasoning.

24.6% percentile inside its fair comparison set

79.5%Raw benchmark value

Spatial

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #52 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 96%
Percentile: 41.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: spatial. Category: Reasoning.

41.5% percentile inside its fair comparison set

96%Raw benchmark value

Logic with navigation

LB · Reasoning / math / science · Objective

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #63 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 62%
Percentile: 6.2%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: logic_with_navigation. Category: Reasoning.

6.2% percentile inside its fair comparison set

62%Raw benchmark value

Professional reasoning30 benchmarks55.8%

Text Arena · Expert

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #28 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,511
Percentile: 91.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: expert. Source rank: #29. Votes: 2442. Organization: zai. License: MIT.

91.7% percentile inside its fair comparison set

1,511Raw benchmark valueCI 1,498 - 1,523

Text Arena · Industry Business And Management And Financial Operations

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #44 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,467
Percentile: 88.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_business_and_management_and_financial_operations. Source rank: #46. Votes: 4388. Organization: zai. License: MIT.

88.3% percentile inside its fair comparison set

1,467Raw benchmark valueCI 1,457 - 1,476

Text Arena · Industry Entertainment And Sports And Media

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #50 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,439
Percentile: 86.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_entertainment_and_sports_and_media. Source rank: #54. Votes: 5762. Organization: zai. License: MIT.

86.9% percentile inside its fair comparison set

1,439Raw benchmark valueCI 1,430 - 1,447

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #51 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,470
Percentile: 85.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_legal_and_government. Source rank: #56. Votes: 1927. Organization: zai. License: MIT.

85.7% percentile inside its fair comparison set

1,470Raw benchmark valueCI 1,456 - 1,484

Text Arena · Industry Life And Physical And Social Science

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,492
Percentile: 91.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_life_and_physical_and_social_science. Source rank: #34. Votes: 3729. Organization: zai. License: MIT.

91.4% percentile inside its fair comparison set

1,492Raw benchmark valueCI 1,482 - 1,502

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #24 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,495
Percentile: 93.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_mathematical. Source rank: #25. Votes: 1301. Organization: zai. License: MIT.

93.5% percentile inside its fair comparison set

1,495Raw benchmark valueCI 1,478 - 1,511

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #46 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,485
Percentile: 87%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_medicine_and_healthcare. Source rank: #47. Votes: 1500. Organization: zai. License: MIT.

87% percentile inside its fair comparison set

1,485Raw benchmark valueCI 1,469 - 1,500

Text Arena · Industry Software And It Services

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #26 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,514
Percentile: 93.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_software_and_it_services. Source rank: #26. Votes: 8872. Organization: zai. License: MIT.

93.4% percentile inside its fair comparison set

1,514Raw benchmark valueCI 1,507 - 1,521

Text Arena · Industry Writing And Literature And Language

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #53 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,448
Percentile: 86.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_writing_and_literature_and_language. Source rank: #58. Votes: 6316. Organization: zai. License: MIT.

86.1% percentile inside its fair comparison set

1,448Raw benchmark valueCI 1,440 - 1,456

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #19 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,512
Percentile: 94.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: expert. Source rank: #19. Votes: 2442. Organization: zai. License: MIT.

94.5% percentile inside its fair comparison set

1,512Raw benchmark valueCI 1,500 - 1,525

Text Arena · Industry Business And Management And Financial Operations · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,458
Percentile: 91.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_business_and_management_and_financial_operations. Source rank: #34. Votes: 4388. Organization: zai. License: MIT.

91.3% percentile inside its fair comparison set

1,458Raw benchmark valueCI 1,449 - 1,468

Text Arena · Industry Entertainment And Sports And Media · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #32 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,442
Percentile: 91.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_entertainment_and_sports_and_media. Source rank: #34. Votes: 5762. Organization: zai. License: MIT.

91.7% percentile inside its fair comparison set

1,442Raw benchmark valueCI 1,433 - 1,451

Text Arena · Industry Legal And Government · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #35 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,473
Percentile: 90.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_legal_and_government. Source rank: #37. Votes: 1927. Organization: zai. License: MIT.

90.3% percentile inside its fair comparison set

1,473Raw benchmark valueCI 1,459 - 1,487

Text Arena · Industry Life And Physical And Social Science · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #25 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,487
Percentile: 93.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_life_and_physical_and_social_science. Source rank: #26. Votes: 3729. Organization: zai. License: MIT.

93.6% percentile inside its fair comparison set

1,487Raw benchmark valueCI 1,477 - 1,496

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #20 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,496
Percentile: 94.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_mathematical. Source rank: #20. Votes: 1301. Organization: zai. License: MIT.

94.7% percentile inside its fair comparison set

1,496Raw benchmark valueCI 1,479 - 1,512

Text Arena · Industry Medicine And Healthcare · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,470
Percentile: 90.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_medicine_and_healthcare. Source rank: #34. Votes: 1500. Organization: zai. License: MIT.

90.8% percentile inside its fair comparison set

1,470Raw benchmark valueCI 1,455 - 1,486

Text Arena · Industry Software And It Services · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,500
Percentile: 95.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_software_and_it_services. Source rank: #18. Votes: 8872. Organization: zai. License: MIT.

95.5% percentile inside its fair comparison set

1,500Raw benchmark valueCI 1,493 - 1,507

Text Arena · Industry Writing And Literature And Language · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #37 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,452
Percentile: 90.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: industry_writing_and_literature_and_language. Source rank: #39. Votes: 6316. Organization: zai. License: MIT.

90.4% percentile inside its fair comparison set

1,452Raw benchmark valueCI 1,444 - 1,460

Legal Research Bench

VALS-AI · Professional reasoning · Objective

Applied legal research tasks.

Rank #15 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 45.2%
Percentile: 79.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_research; provider: Zhipu AI.

79.4% percentile inside its fair comparison set

45.2%Raw benchmark valueCI 38.4% - 52%

Harvey's Legal Agent Benchmark

VALS-AI · Professional reasoning · Objective

Completing legal work with documents, spreadsheets, presentations, and file-system tools.

Rank #25 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 6.7%
Percentile: 68.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Zhipu AI.

68.1% percentile inside its fair comparison set

6.7%Raw benchmark valueCI 3.4% - 9.9%

LegalBench

VALS-AI · Professional reasoning · Objective

Academic legal reasoning tasks.

Rank #44 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 83.9%
Percentile: 68.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

68.1% percentile inside its fair comparison set

83.9%Raw benchmark valueCI 83.1% - 84.8%

Finance Agent v2

VALS-AI · Professional reasoning · Objective

Core financial analyst tasks for agentic models.

Rank #11 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 57.9%
Percentile: 85.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: fabv2; provider: Fireworks AI.

85.5% percentile inside its fair comparison set

57.9%Raw benchmark valueCI 53.8% - 61.9%

MedScribe

VALS-AI · Professional reasoning · Objective

Administrative documentation support for doctors.

Rank #7 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 88.9%
Percentile: 93.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: medscribe; provider: Zhipu AI.

93.8% percentile inside its fair comparison set

88.9%Raw benchmark valueCI 85.2% - 92.7%

SkillsBench

VALS-AI · Professional reasoning · Objective

Applied professional skills tasks.

Rank #29 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 40.2%
Percentile: 15.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: skillsbench; provider: Fireworks AI.

15.2% percentile inside its fair comparison set

40.2%Raw benchmark valueCI 31.5% - 48.8%

TaxEval v2

VALS-AI · Professional reasoning · Objective

Answer quality on tax questions and responses.

Rank #16 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 75.6%
Percentile: 88.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: tax_eval_v2; provider: Zhipu AI.

88.4% percentile inside its fair comparison set

75.6%Raw benchmark valueCI 74% - 77.2%

Data analysis

LB · Professional reasoning · Objective

Structured data manipulation and table reasoning accuracy.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 76.4%
Percentile: 50.8%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category: Data Analysis. Tasks scored: 3.

50.8% percentile inside its fair comparison set

76.4%Raw benchmark value

Overall

LB · Professional reasoning · Objective

Average objective performance across LiveBench's current public category mix.

Rank #52 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 71.6%
Percentile: 21.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Category averages included: 7.

21.5% percentile inside its fair comparison set

71.6%Raw benchmark value

Consecutive events

LB · Professional reasoning · Objective

Objective consecutive events score in LiveBench.

Rank #33 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 84.3%
Percentile: 50.8%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: consecutive_events. Category: Data Analysis.

50.8% percentile inside its fair comparison set

84.3%Raw benchmark value

Table join

LB · Professional reasoning · Objective

Objective table join score in LiveBench.

Rank #28 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 48.8%
Percentile: 58.5%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: tablejoin. Category: Data Analysis.

58.5% percentile inside its fair comparison set

48.8%Raw benchmark value

Table reformat

LB · Professional reasoning · Objective

Objective table reformat score in LiveBench.

Rank #59 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: LiveBench
Raw value: 96.1%
Percentile: 20%
Last updated: stale
Eligibility: headline eligible

Derived from the official LiveBench website leaderboard table. Task: tablereformat. Category: Data Analysis.

20% percentile inside its fair comparison set

96.1%Raw benchmark value

Long context1 benchmark88.6%

Long Context Reasoning

AA · Long context · Objective

It checks whether long-context claims survive contact with retrieval, memory, or long-document tasks.

Rank #53 · Source label: GLM 5.3 Flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 80%
Percentile: 88.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `lcr`.

88.6% percentile inside its fair comparison set

80%Raw benchmark value

Vision understanding15 benchmarks77.4%

Vision Arena

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #27 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,278
Percentile: 81.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: overall. Source rank: #30. Votes: 5479. Organization: zai. License: MIT.

81.3% percentile inside its fair comparison set

1,278Raw benchmark valueCI 1,269 - 1,288

Vision Arena · Creative Writing Vision

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #28 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,259
Percentile: 67.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: creative_writing_vision. Source rank: #32. Votes: 493. Organization: zai. License: MIT.

67.1% percentile inside its fair comparison set

1,259Raw benchmark valueCI 1,232 - 1,287

Vision Arena · Diagram

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,313
Percentile: 82.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: diagram. Source rank: #20. Votes: 1299. Organization: zai. License: MIT.

82.8% percentile inside its fair comparison set

1,313Raw benchmark valueCI 1,296 - 1,330

Vision Arena · English

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #30 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,272
Percentile: 79%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: english. Source rank: #33. Votes: 1731. Organization: zai. License: MIT.

79% percentile inside its fair comparison set

1,272Raw benchmark valueCI 1,257 - 1,287

Vision Arena · Homework

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #11 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,331
Percentile: 89.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: homework. Source rank: #12. Votes: 616. Organization: zai. License: MIT.

89.6% percentile inside its fair comparison set

1,331Raw benchmark valueCI 1,307 - 1,355

Vision Arena · Humor

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #31 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,251
Percentile: 58.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: humor. Source rank: #35. Votes: 242. Organization: zai. License: MIT.

58.9% percentile inside its fair comparison set

1,251Raw benchmark valueCI 1,213 - 1,290

Vision Arena · Ocr

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #28 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,291
Percentile: 73%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: ocr. Source rank: #30. Votes: 3682. Organization: zai. License: MIT.

73% percentile inside its fair comparison set

1,291Raw benchmark valueCI 1,280 - 1,301

Vision Arena · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #17 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,301
Percentile: 88.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: overall. Source rank: #18. Votes: 5479. Organization: zai. License: MIT.

88.5% percentile inside its fair comparison set

1,301Raw benchmark valueCI 1,292 - 1,310

Vision Arena · Creative Writing Vision · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #21 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,286
Percentile: 75.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: creative_writing_vision. Source rank: #23. Votes: 493. Organization: zai. License: MIT.

75.6% percentile inside its fair comparison set

1,286Raw benchmark valueCI 1,258 - 1,313

Vision Arena · Diagram · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #15 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,323
Percentile: 85.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: diagram. Source rank: #16. Votes: 1299. Organization: zai. License: MIT.

85.9% percentile inside its fair comparison set

1,323Raw benchmark valueCI 1,306 - 1,340

Vision Arena · English · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #15 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,297
Percentile: 89.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: english. Source rank: #16. Votes: 1731. Organization: zai. License: MIT.

89.9% percentile inside its fair comparison set

1,297Raw benchmark valueCI 1,282 - 1,312

Vision Arena · Homework · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #7 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,343
Percentile: 93.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: homework. Source rank: #7. Votes: 616. Organization: zai. License: MIT.

93.8% percentile inside its fair comparison set

1,343Raw benchmark valueCI 1,319 - 1,367

Vision Arena · Humor · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #25 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,274
Percentile: 67.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: humor. Source rank: #28. Votes: 242. Organization: zai. License: MIT.

67.1% percentile inside its fair comparison set

1,274Raw benchmark valueCI 1,236 - 1,313

Vision Arena · Ocr · No Style Control

AR · Vision understanding · Human

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #18 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,308
Percentile: 83%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: ocr. Source rank: #19. Votes: 3682. Organization: zai. License: MIT.

83% percentile inside its fair comparison set

1,308Raw benchmark valueCI 1,297 - 1,318

MMMU Pro

VALS-AI · Vision understanding · Objective

It is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.

Rank #23 · Source label: zai/glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 86%
Percentile: 73.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Vals AI BenchmarkView overall scores. Vals slug: mmmu; provider: Zhipu AI.

73.5% percentile inside its fair comparison set

86%Raw benchmark valueCI 84.4% - 87.6%

Multilingual20 benchmarksn/a

Text Arena · Chinese

AR · Multilingual · Human

Observed user preference in Arena's Text Arena chinese leaderboard.

Rank #21 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,529
Percentile: 94.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: chinese. Source rank: #21. Votes: 1648. Organization: zai. License: MIT.

94.3% percentile inside its fair comparison set

1,529Raw benchmark valueCI 1,513 - 1,544

Text Arena · French

AR · Multilingual · Human

Observed user preference in Arena's Text Arena french leaderboard.

Rank #15 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,507
Percentile: 94.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: french. Source rank: #15. Votes: 765. Organization: zai. License: MIT.

94.6% percentile inside its fair comparison set

1,507Raw benchmark valueCI 1,485 - 1,530

Text Arena · German

AR · Multilingual · Human

Observed user preference in Arena's Text Arena german leaderboard.

Rank #44 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,465
Percentile: 84.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: german. Source rank: #48. Votes: 395. Organization: zai. License: MIT.

84.2% percentile inside its fair comparison set

1,465Raw benchmark valueCI 1,435 - 1,494

Text Arena · Japanese

AR · Multilingual · Human

Observed user preference in Arena's Text Arena japanese leaderboard.

Rank #45 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,433
Percentile: 81.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: japanese. Source rank: #50. Votes: 408. Organization: zai. License: MIT.

81.9% percentile inside its fair comparison set

1,433Raw benchmark valueCI 1,402 - 1,464

Text Arena · Korean

AR · Multilingual · Human

Observed user preference in Arena's Text Arena korean leaderboard.

Rank #30 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,438
Percentile: 88.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: korean. Source rank: #31. Votes: 397. Organization: zai. License: MIT.

88.1% percentile inside its fair comparison set

1,438Raw benchmark valueCI 1,408 - 1,469

Text Arena · Non English

AR · Multilingual · Human

Observed user preference in Arena's Text Arena non_english leaderboard.

Rank #35 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,463
Percentile: 91%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: non_english. Source rank: #37. Votes: 14422. Organization: zai. License: MIT.

91% percentile inside its fair comparison set

1,463Raw benchmark valueCI 1,457 - 1,469

Text Arena · Polish

AR · Multilingual · Human

Observed user preference in Arena's Text Arena polish leaderboard.

Rank #60 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,447
Percentile: 70.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: polish. Source rank: #73. Votes: 370. Organization: zai. License: MIT.

70.4% percentile inside its fair comparison set

1,447Raw benchmark valueCI 1,417 - 1,476

Text Arena · Russian

AR · Multilingual · Human

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #36 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,475
Percentile: 89.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: russian. Source rank: #40. Votes: 3160. Organization: zai. License: MIT.

89.7% percentile inside its fair comparison set

1,475Raw benchmark valueCI 1,464 - 1,486

Text Arena · Spanish

AR · Multilingual · Human

Observed user preference in Arena's Text Arena spanish leaderboard.

Rank #37 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,466
Percentile: 86.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: spanish. Source rank: #39. Votes: 792. Organization: zai. License: MIT.

86.1% percentile inside its fair comparison set

1,466Raw benchmark valueCI 1,444 - 1,487

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena chinese leaderboard.

Rank #23 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,528
Percentile: 93.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: chinese. Source rank: #23. Votes: 1648. Organization: zai. License: MIT.

93.8% percentile inside its fair comparison set

1,528Raw benchmark valueCI 1,513 - 1,543

Text Arena · French · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena french leaderboard.

Rank #13 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,498
Percentile: 95.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: french. Source rank: #13. Votes: 765. Organization: zai. License: MIT.

95.3% percentile inside its fair comparison set

1,498Raw benchmark valueCI 1,476 - 1,521

Text Arena · German · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena german leaderboard.

Rank #37 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,458
Percentile: 86.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: german. Source rank: #40. Votes: 395. Organization: zai. License: MIT.

86.8% percentile inside its fair comparison set

1,458Raw benchmark valueCI 1,428 - 1,488

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena japanese leaderboard.

Rank #40 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,425
Percentile: 84%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: japanese. Source rank: #43. Votes: 408. Organization: zai. License: MIT.

84% percentile inside its fair comparison set

1,425Raw benchmark valueCI 1,394 - 1,456

Text Arena · Korean · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena korean leaderboard.

Rank #25 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,439
Percentile: 90.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: korean. Source rank: #27. Votes: 397. Organization: zai. License: MIT.

90.2% percentile inside its fair comparison set

1,439Raw benchmark valueCI 1,408 - 1,469

Text Arena · Non English · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena non_english leaderboard.

Rank #28 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,461
Percentile: 92.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: non_english. Source rank: #29. Votes: 14422. Organization: zai. License: MIT.

92.8% percentile inside its fair comparison set

1,461Raw benchmark valueCI 1,455 - 1,467

Text Arena · Polish · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena polish leaderboard.

Rank #55 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,439
Percentile: 72.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: polish. Source rank: #62. Votes: 370. Organization: zai. License: MIT.

72.9% percentile inside its fair comparison set

1,439Raw benchmark valueCI 1,409 - 1,469

Text Arena · Russian · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #29 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,471
Percentile: 91.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: russian. Source rank: #31. Votes: 3160. Organization: zai. License: MIT.

91.8% percentile inside its fair comparison set

1,471Raw benchmark valueCI 1,460 - 1,482

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena spanish leaderboard.

Rank #27 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,469
Percentile: 90%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: spanish. Source rank: #28. Votes: 792. Organization: zai. License: MIT.

90% percentile inside its fair comparison set

1,469Raw benchmark valueCI 1,448 - 1,490

Vision Arena · Chinese

AR · Multilingual · Human

Observed user preference in Arena's Vision Arena chinese leaderboard.

Rank #8 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,349
Percentile: 93.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: chinese. Source rank: #8. Votes: 231. Organization: zai. License: MIT.

93.3% percentile inside its fair comparison set

1,349Raw benchmark valueCI 1,311 - 1,387

Vision Arena · Chinese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Vision Arena chinese leaderboard.

Rank #5 · Source label: glm-5.3-flash

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,373
Percentile: 96.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `glm-5.3-flash`. Category: chinese. Source rank: #6. Votes: 231. Organization: zai. License: MIT.

96.2% percentile inside its fair comparison set

1,373Raw benchmark valueCI 1,335 - 1,411

Source links and registry checks

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Arena

Oct 3, 2026

source →

official

Artificial Analysis

Oct 3, 2026

source →

official

LiveBench

Oct 3, 2026

source →

glm-5.3-flash

Current snapshot.

Source-linked scores by benchmark

Source links and registry checks

Loading model evidence.

glm-5.3-flash

Current snapshot.

Source-linked scores by benchmark

Source links and registry checks