Model profile · Qwen

Qwen3.5 27B

Open weightsbudget · registry tag 2026 open incremental

Thin verified coverage

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

Visible coverage: 13%
Verified coverage: 13%
Spread: 60.6%
Last verified: Oct 3, 2026

60%capability fit

textcodevisiondocument6 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 / text27 benchmarks60.8%

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 #169 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 19
Percentile: 69.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.

69.7% percentile inside its fair comparison set

19Raw 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 #309 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 15.7%
Percentile: 32%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.

32% percentile inside its fair comparison set

15.7%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 #234 · Source label: Qwen3.5 27B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 18.5%
Percentile: 48.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.

48.6% percentile inside its fair comparison set

18.5%Raw benchmark value

IFBench

AA · Chat / text · Objective

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

Rank #126 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 46.9%
Percentile: 64.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `ifbench`.

64.1% percentile inside its fair comparison set

46.9%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 #133 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $0.3 /1M input tokens
Percentile: 63.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.

63.8% percentile inside its fair comparison set

$0.3 /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 #177 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $2.4 /1M output tokens
Percentile: 50.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.

50.6% percentile inside its fair comparison set

$2.4 /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 #153 · Source label: Qwen3.5 27B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 75.4 tokens/s
Percentile: 42%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

42% percentile inside its fair comparison set

75.4 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 #181 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 5.71s
Percentile: 31.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.

31.8% percentile inside its fair comparison set

5.71sRaw 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 #181 · Source label: Qwen3.5 27B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 32.19s
Percentile: 31.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.

31.3% percentile inside its fair comparison set

32.19sRaw 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 #135 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,409
Percentile: 64.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: overall. Source rank: #150. Votes: 28612. Organization: alibaba. License: Apache 2.0.

64.4% percentile inside its fair comparison set

1,409Raw benchmark valueCI 1,404 - 1,413

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 #148 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,358
Percentile: 60.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: creative_writing. Source rank: #168. Votes: 4297. Organization: alibaba. License: Apache 2.0.

60.7% percentile inside its fair comparison set

1,358Raw benchmark valueCI 1,349 - 1,367

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 #119 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,428
Percentile: 68.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: english. Source rank: #133. Votes: 12729. Organization: alibaba. License: Apache 2.0.

68.6% percentile inside its fair comparison set

1,428Raw benchmark valueCI 1,422 - 1,434

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 #136 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,393
Percentile: 64.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: exclude_ties. Source rank: #151. Votes: 20882. Organization: alibaba. License: Apache 2.0.

64.1% percentile inside its fair comparison set

1,393Raw benchmark valueCI 1,387 - 1,399

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 #132 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,428
Percentile: 65.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: hard_prompts. Source rank: #147. Votes: 18070. Organization: alibaba. License: Apache 2.0.

65.2% percentile inside its fair comparison set

1,428Raw benchmark valueCI 1,422 - 1,433

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 #120 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,444
Percentile: 68.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: hard_prompts_english. Source rank: #134. Votes: 8339. Organization: alibaba. License: Apache 2.0.

68.2% percentile inside its fair comparison set

1,444Raw benchmark valueCI 1,437 - 1,451

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 #127 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,401
Percentile: 66.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: instruction_following. Source rank: #142. Votes: 9207. Organization: alibaba. License: Apache 2.0.

66.5% percentile inside its fair comparison set

1,401Raw benchmark valueCI 1,394 - 1,407

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 #120 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,422
Percentile: 66.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: longer_query. Source rank: #135. Votes: 11035. Organization: alibaba. License: Apache 2.0.

66.5% percentile inside its fair comparison set

1,422Raw benchmark valueCI 1,416 - 1,429

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 #127 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,414
Percentile: 66.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: multi_turn. Source rank: #142. Votes: 4891. Organization: alibaba. License: Apache 2.0.

66.3% percentile inside its fair comparison set

1,414Raw benchmark valueCI 1,405 - 1,423

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 #127 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,409
Percentile: 66.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: overall. Source rank: #139. Votes: 28612. Organization: alibaba. License: Apache 2.0.

66.5% percentile inside its fair comparison set

1,409Raw benchmark valueCI 1,405 - 1,413

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 #135 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 64.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: creative_writing. Source rank: #151. Votes: 4297. Organization: alibaba. License: Apache 2.0.

64.2% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,353 - 1,372

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 #117 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,425
Percentile: 69.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: english. Source rank: #126. Votes: 12729. Organization: alibaba. License: Apache 2.0.

69.1% percentile inside its fair comparison set

1,425Raw benchmark valueCI 1,419 - 1,431

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 #127 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,392
Percentile: 66.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: exclude_ties. Source rank: #139. Votes: 20882. Organization: alibaba. License: Apache 2.0.

66.5% percentile inside its fair comparison set

1,392Raw benchmark valueCI 1,386 - 1,397

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 #128 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,414
Percentile: 66.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: hard_prompts. Source rank: #140. Votes: 18070. Organization: alibaba. License: Apache 2.0.

66.2% percentile inside its fair comparison set

1,414Raw benchmark valueCI 1,409 - 1,420

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 #118 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,428
Percentile: 68.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: hard_prompts_english. Source rank: #129. Votes: 8339. Organization: alibaba. License: Apache 2.0.

68.7% percentile inside its fair comparison set

1,428Raw benchmark valueCI 1,421 - 1,435

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 #124 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,393
Percentile: 67.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: instruction_following. Source rank: #137. Votes: 9207. Organization: alibaba. License: Apache 2.0.

67.3% percentile inside its fair comparison set

1,393Raw benchmark valueCI 1,386 - 1,399

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 #113 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,413
Percentile: 68.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: longer_query. Source rank: #121. Votes: 11035. Organization: alibaba. License: Apache 2.0.

68.5% percentile inside its fair comparison set

1,413Raw benchmark valueCI 1,407 - 1,419

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 #121 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,410
Percentile: 67.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: multi_turn. Source rank: #134. Votes: 4891. Organization: alibaba. License: Apache 2.0.

67.9% percentile inside its fair comparison set

1,410Raw benchmark valueCI 1,401 - 1,419

Coding6 benchmarks52.7%

Terminal-Bench Hard

AA · Coding · Objective

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

Rank #61 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 31.8%
Percentile: 82.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `terminalbenchHard`.

82.5% percentile inside its fair comparison set

31.8%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 #137 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,452
Percentile: 63.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: coding. Source rank: #152. Votes: 7939. Organization: alibaba. License: Apache 2.0.

63.3% percentile inside its fair comparison set

1,452Raw benchmark valueCI 1,444 - 1,459

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 #124 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,427
Percentile: 66.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: coding. Source rank: #136. Votes: 7939. Organization: alibaba. License: Apache 2.0.

66.8% percentile inside its fair comparison set

1,427Raw benchmark valueCI 1,420 - 1,434

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 #99 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,357
Percentile: 22.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: overall. Source rank: #104. Votes: 7794. Organization: alibaba. License: Apache 2.0.

22.8% percentile inside its fair comparison set

1,357Raw benchmark valueCI 1,349 - 1,365

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 #90 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,386
Percentile: 28.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: webdev-html. Source rank: #95. Votes: 940. Organization: alibaba. License: Apache 2.0.

28.8% percentile inside its fair comparison set

1,386Raw benchmark valueCI 1,366 - 1,407

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 #96 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,346
Percentile: 15.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: webdev-react. Source rank: #103. Votes: 6763. Organization: alibaba. License: Apache 2.0.

15.9% percentile inside its fair comparison set

1,346Raw benchmark valueCI 1,337 - 1,354

Reasoning / math / science5 benchmarks69.1%

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 #180 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 13.9%
Percentile: 66.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `hle`.

66.2% percentile inside its fair comparison set

13.9%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 #92 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 84.2%
Percentile: 81.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `gpqa`.

81.5% percentile inside its fair comparison set

84.2%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 #189 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 0.3%
Percentile: 59.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `critpt`.

59.5% percentile inside its fair comparison set

0.3%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 #107 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,423
Percentile: 70.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: math. Source rank: #119. Votes: 1744. Organization: alibaba. License: Apache 2.0.

70.5% percentile inside its fair comparison set

1,423Raw benchmark valueCI 1,409 - 1,437

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 #92 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,429
Percentile: 74.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: math. Source rank: #100. Votes: 1744. Organization: alibaba. License: Apache 2.0.

74.7% percentile inside its fair comparison set

1,429Raw benchmark valueCI 1,415 - 1,443

Professional reasoning18 benchmarksn/a

Text Arena · Expert

AR · Professional reasoning · Human

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

Rank #116 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,441
Percentile: 64.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: expert. Source rank: #129. Votes: 2646. Organization: alibaba. License: Apache 2.0.

64.8% percentile inside its fair comparison set

1,441Raw benchmark valueCI 1,429 - 1,453

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 #132 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,408
Percentile: 64.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_business_and_management_and_financial_operations. Source rank: #146. Votes: 5602. Organization: alibaba. License: Apache 2.0.

64.5% percentile inside its fair comparison set

1,408Raw benchmark valueCI 1,400 - 1,416

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 #143 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,363
Percentile: 62%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_entertainment_and_sports_and_media. Source rank: #162. Votes: 5373. Organization: alibaba. License: Apache 2.0.

62% percentile inside its fair comparison set

1,363Raw benchmark valueCI 1,354 - 1,371

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

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

Rank #140 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,406
Percentile: 60.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_legal_and_government. Source rank: #157. Votes: 2184. Organization: alibaba. License: Apache 2.0.

60.2% percentile inside its fair comparison set

1,406Raw benchmark valueCI 1,393 - 1,419

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 #111 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,441
Percentile: 70.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_life_and_physical_and_social_science. Source rank: #124. Votes: 4772. Organization: alibaba. License: Apache 2.0.

70.6% percentile inside its fair comparison set

1,441Raw benchmark valueCI 1,432 - 1,449

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

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

Rank #113 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,428
Percentile: 68.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_mathematical. Source rank: #124. Votes: 1681. Organization: alibaba. License: Apache 2.0.

68.5% percentile inside its fair comparison set

1,428Raw benchmark valueCI 1,413 - 1,443

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

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

Rank #133 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,431
Percentile: 61.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_medicine_and_healthcare. Source rank: #148. Votes: 2046. Organization: alibaba. License: Apache 2.0.

61.8% percentile inside its fair comparison set

1,431Raw benchmark valueCI 1,417 - 1,444

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 #129 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,446
Percentile: 66%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_software_and_it_services. Source rank: #143. Votes: 11345. Organization: alibaba. License: Apache 2.0.

66% percentile inside its fair comparison set

1,446Raw benchmark valueCI 1,440 - 1,453

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 #139 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,380
Percentile: 63.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_writing_and_literature_and_language. Source rank: #157. Votes: 6479. Organization: alibaba. License: Apache 2.0.

63.2% percentile inside its fair comparison set

1,380Raw benchmark valueCI 1,373 - 1,388

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

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

Rank #108 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,428
Percentile: 67.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: expert. Source rank: #116. Votes: 2646. Organization: alibaba. License: Apache 2.0.

67.3% percentile inside its fair comparison set

1,428Raw benchmark valueCI 1,416 - 1,440

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 #126 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,398
Percentile: 66.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_business_and_management_and_financial_operations. Source rank: #137. Votes: 5602. Organization: alibaba. License: Apache 2.0.

66.1% percentile inside its fair comparison set

1,398Raw benchmark valueCI 1,390 - 1,407

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 #135 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,363
Percentile: 64.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_entertainment_and_sports_and_media. Source rank: #150. Votes: 5373. Organization: alibaba. License: Apache 2.0.

64.2% percentile inside its fair comparison set

1,363Raw benchmark valueCI 1,355 - 1,372

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 #135 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,405
Percentile: 61.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_legal_and_government. Source rank: #149. Votes: 2184. Organization: alibaba. License: Apache 2.0.

61.6% percentile inside its fair comparison set

1,405Raw benchmark valueCI 1,392 - 1,418

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 #103 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,436
Percentile: 72.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_life_and_physical_and_social_science. Source rank: #111. Votes: 4772. Organization: alibaba. License: Apache 2.0.

72.7% percentile inside its fair comparison set

1,436Raw benchmark valueCI 1,427 - 1,445

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

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

Rank #107 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,429
Percentile: 70.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_mathematical. Source rank: #114. Votes: 1681. Organization: alibaba. License: Apache 2.0.

70.2% percentile inside its fair comparison set

1,429Raw benchmark valueCI 1,414 - 1,443

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 #114 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,424
Percentile: 67.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_medicine_and_healthcare. Source rank: #123. Votes: 2046. Organization: alibaba. License: Apache 2.0.

67.3% percentile inside its fair comparison set

1,424Raw benchmark valueCI 1,411 - 1,437

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 #121 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,431
Percentile: 68.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_software_and_it_services. Source rank: #133. Votes: 11345. Organization: alibaba. License: Apache 2.0.

68.1% percentile inside its fair comparison set

1,431Raw benchmark valueCI 1,425 - 1,437

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 #132 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,381
Percentile: 65.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_writing_and_literature_and_language. Source rank: #146. Votes: 6479. Organization: alibaba. License: Apache 2.0.

65.1% percentile inside its fair comparison set

1,381Raw benchmark valueCI 1,373 - 1,388

Search / tool use1 benchmark83.5%

Tau2-Bench Telecom

AA · Search / tool use · Objective

It matters when the model must browse, call tools, and recover useful answers from external systems.

Rank #57 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 87.1%
Percentile: 83.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `tau2`.

83.5% percentile inside its fair comparison set

87.1%Raw benchmark value

Long context1 benchmark61.3%

Long Context Reasoning

AA · Long context · Objective

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

Rank #178 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 64%
Percentile: 61.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `lcr`.

61.3% percentile inside its fair comparison set

64%Raw benchmark value

Vision understanding19 benchmarks54.4%

MMMU-Pro

AA · Vision understanding · Objective

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

Rank #95 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 70%
Percentile: 54.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

54.8% percentile inside its fair comparison set

70%Raw benchmark value

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 #65 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,216
Percentile: 54%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: overall. Source rank: #73. Votes: 28993. Organization: alibaba. License: Apache 2.0.

54% percentile inside its fair comparison set

1,216Raw benchmark valueCI 1,211 - 1,222

Vision Arena · Captioning

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: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,097
Percentile: 6.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: captioning. Source rank: #32. Votes: 61. Organization: alibaba. License: Apache 2.0.

6.5% percentile inside its fair comparison set

1,097Raw benchmark valueCI 1,015 - 1,179

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 #63 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,198
Percentile: 24.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: creative_writing_vision. Source rank: #72. Votes: 2014. Organization: alibaba. License: Apache 2.0.

24.4% percentile inside its fair comparison set

1,198Raw benchmark valueCI 1,184 - 1,212

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 #66 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,239
Percentile: 34.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: diagram. Source rank: #75. Votes: 7586. Organization: alibaba. License: Apache 2.0.

34.3% percentile inside its fair comparison set

1,239Raw benchmark valueCI 1,231 - 1,247

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 #63 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,221
Percentile: 55.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: english. Source rank: #71. Votes: 11319. Organization: alibaba. License: Apache 2.0.

55.1% percentile inside its fair comparison set

1,221Raw benchmark valueCI 1,214 - 1,229

Vision Arena · Entity Recognition

AR · Vision understanding · Human

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

Rank #36 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,144
Percentile: 12.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: entity_recognition. Source rank: #45. Votes: 180. Organization: alibaba. License: Apache 2.0.

12.5% percentile inside its fair comparison set

1,144Raw benchmark valueCI 1,101 - 1,186

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 #68 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,241
Percentile: 30.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: homework. Source rank: #77. Votes: 3651. Organization: alibaba. License: Apache 2.0.

30.2% percentile inside its fair comparison set

1,241Raw benchmark valueCI 1,231 - 1,252

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 #57 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,192
Percentile: 23.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: humor. Source rank: #66. Votes: 1130. Organization: alibaba. License: Apache 2.0.

23.3% percentile inside its fair comparison set

1,192Raw benchmark valueCI 1,174 - 1,210

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 #65 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,228
Percentile: 36%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: ocr. Source rank: #74. Votes: 20452. Organization: alibaba. License: Apache 2.0.

36% percentile inside its fair comparison set

1,228Raw benchmark valueCI 1,223 - 1,234

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 #63 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,240
Percentile: 55.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: overall. Source rank: #69. Votes: 28993. Organization: alibaba. License: Apache 2.0.

55.4% percentile inside its fair comparison set

1,240Raw benchmark valueCI 1,235 - 1,246

Vision Arena · Captioning · 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 #26 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,172
Percentile: 19.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: captioning. Source rank: #27. Votes: 61. Organization: alibaba. License: Apache 2.0.

19.4% percentile inside its fair comparison set

1,172Raw benchmark valueCI 1,091 - 1,253

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 #58 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,224
Percentile: 30.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: creative_writing_vision. Source rank: #66. Votes: 2014. Organization: alibaba. License: Apache 2.0.

30.5% percentile inside its fair comparison set

1,224Raw benchmark valueCI 1,210 - 1,238

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 #65 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,248
Percentile: 35.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: diagram. Source rank: #71. Votes: 7586. Organization: alibaba. License: Apache 2.0.

35.4% percentile inside its fair comparison set

1,248Raw benchmark valueCI 1,240 - 1,256

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 #59 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,245
Percentile: 58%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: english. Source rank: #64. Votes: 11319. Organization: alibaba. License: Apache 2.0.

58% percentile inside its fair comparison set

1,245Raw benchmark valueCI 1,238 - 1,253

Vision Arena · Entity Recognition · 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 #30 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,205
Percentile: 27.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: entity_recognition. Source rank: #37. Votes: 180. Organization: alibaba. License: Apache 2.0.

27.5% percentile inside its fair comparison set

1,205Raw benchmark valueCI 1,162 - 1,247

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 #59 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,254
Percentile: 39.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: homework. Source rank: #66. Votes: 3651. Organization: alibaba. License: Apache 2.0.

39.6% percentile inside its fair comparison set

1,254Raw benchmark valueCI 1,244 - 1,265

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 #50 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,223
Percentile: 32.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: humor. Source rank: #58. Votes: 1130. Organization: alibaba. License: Apache 2.0.

32.9% percentile inside its fair comparison set

1,223Raw benchmark valueCI 1,205 - 1,242

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 #64 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,245
Percentile: 37%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: ocr. Source rank: #70. Votes: 20452. Organization: alibaba. License: Apache 2.0.

37% percentile inside its fair comparison set

1,245Raw benchmark valueCI 1,240 - 1,251

Multilingual20 benchmarksn/a

Text Arena · Chinese

AR · Multilingual · Human

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

Rank #98 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,472
Percentile: 72.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: chinese. Source rank: #105. Votes: 2041. Organization: alibaba. License: Apache 2.0.

72.4% percentile inside its fair comparison set

1,472Raw benchmark valueCI 1,459 - 1,486

Text Arena · French

AR · Multilingual · Human

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

Rank #130 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,416
Percentile: 49.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: french. Source rank: #146. Votes: 849. Organization: alibaba. License: Apache 2.0.

49.8% percentile inside its fair comparison set

1,416Raw benchmark valueCI 1,394 - 1,437

Text Arena · German

AR · Multilingual · Human

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

Rank #116 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,391
Percentile: 57.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: german. Source rank: #132. Votes: 452. Organization: alibaba. License: Apache 2.0.

57.7% percentile inside its fair comparison set

1,391Raw benchmark valueCI 1,363 - 1,418

Text Arena · Japanese

AR · Multilingual · Human

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

Rank #117 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,338
Percentile: 52.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: japanese. Source rank: #135. Votes: 224. Organization: alibaba. License: Apache 2.0.

52.3% percentile inside its fair comparison set

1,338Raw benchmark valueCI 1,298 - 1,378

Text Arena · Korean

AR · Multilingual · Human

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

Rank #117 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,350
Percentile: 52.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: korean. Source rank: #133. Votes: 446. Organization: alibaba. License: Apache 2.0.

52.5% percentile inside its fair comparison set

1,350Raw benchmark valueCI 1,322 - 1,379

Text Arena · Non English

AR · Multilingual · Human

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

Rank #140 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,387
Percentile: 63%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: non_english. Source rank: #157. Votes: 15883. Organization: alibaba. License: Apache 2.0.

63% percentile inside its fair comparison set

1,387Raw benchmark valueCI 1,381 - 1,392

Text Arena · Polish

AR · Multilingual · Human

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

Rank #136 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,366
Percentile: 32.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: polish. Source rank: #159. Votes: 538. Organization: alibaba. License: Apache 2.0.

32.2% percentile inside its fair comparison set

1,366Raw benchmark valueCI 1,341 - 1,392

Text Arena · Russian

AR · Multilingual · Human

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

Rank #142 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,390
Percentile: 58.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: russian. Source rank: #160. Votes: 3314. Organization: alibaba. License: Apache 2.0.

58.7% percentile inside its fair comparison set

1,390Raw benchmark valueCI 1,379 - 1,400

Text Arena · Spanish

AR · Multilingual · Human

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

Rank #130 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,398
Percentile: 50.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: spanish. Source rank: #147. Votes: 940. Organization: alibaba. License: Apache 2.0.

50.2% percentile inside its fair comparison set

1,398Raw benchmark valueCI 1,379 - 1,418

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #72 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,478
Percentile: 79.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: chinese. Source rank: #75. Votes: 2041. Organization: alibaba. License: Apache 2.0.

79.8% percentile inside its fair comparison set

1,478Raw benchmark valueCI 1,465 - 1,491

Text Arena · French · No Style Control

AR · Multilingual · Human

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

Rank #123 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,409
Percentile: 52.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: french. Source rank: #136. Votes: 849. Organization: alibaba. License: Apache 2.0.

52.5% percentile inside its fair comparison set

1,409Raw benchmark valueCI 1,388 - 1,430

Text Arena · German · No Style Control

AR · Multilingual · Human

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

Rank #110 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,393
Percentile: 59.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: german. Source rank: #124. Votes: 452. Organization: alibaba. License: Apache 2.0.

59.9% percentile inside its fair comparison set

1,393Raw benchmark valueCI 1,366 - 1,420

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

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

Rank #108 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,345
Percentile: 56%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: japanese. Source rank: #123. Votes: 224. Organization: alibaba. License: Apache 2.0.

56% percentile inside its fair comparison set

1,345Raw benchmark valueCI 1,304 - 1,385

Text Arena · Korean · No Style Control

AR · Multilingual · Human

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

Rank #108 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,358
Percentile: 56.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: korean. Source rank: #122. Votes: 446. Organization: alibaba. License: Apache 2.0.

56.1% percentile inside its fair comparison set

1,358Raw benchmark valueCI 1,329 - 1,387

Text Arena · Non English · No Style Control

AR · Multilingual · Human

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

Rank #133 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,390
Percentile: 64.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: non_english. Source rank: #147. Votes: 15883. Organization: alibaba. License: Apache 2.0.

64.9% percentile inside its fair comparison set

1,390Raw benchmark valueCI 1,384 - 1,395

Text Arena · Polish · No Style Control

AR · Multilingual · Human

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

Rank #123 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,375
Percentile: 38.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: polish. Source rank: #141. Votes: 538. Organization: alibaba. License: Apache 2.0.

38.7% percentile inside its fair comparison set

1,375Raw benchmark valueCI 1,351 - 1,400

Text Arena · Russian · No Style Control

AR · Multilingual · Human

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

Rank #134 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,390
Percentile: 61%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: russian. Source rank: #149. Votes: 3314. Organization: alibaba. License: Apache 2.0.

61% percentile inside its fair comparison set

1,390Raw benchmark valueCI 1,379 - 1,400

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

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

Rank #118 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,407
Percentile: 54.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: spanish. Source rank: #129. Votes: 940. Organization: alibaba. License: Apache 2.0.

54.8% percentile inside its fair comparison set

1,407Raw benchmark valueCI 1,387 - 1,426

Vision Arena · Chinese

AR · Multilingual · Human

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

Rank #52 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,271
Percentile: 51%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: chinese. Source rank: #58. Votes: 1923. Organization: alibaba. License: Apache 2.0.

51% percentile inside its fair comparison set

1,271Raw benchmark valueCI 1,255 - 1,287

Vision Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #45 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,301
Percentile: 57.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: chinese. Source rank: #49. Votes: 1923. Organization: alibaba. License: Apache 2.0.

57.7% percentile inside its fair comparison set

1,301Raw benchmark valueCI 1,285 - 1,317

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

Arena

Oct 3, 2026

source →

official

Artificial Analysis

Oct 3, 2026

source →

Model profile · Qwen

Qwen3.5 27B

Open weightsbudget · registry tag 2026 open incremental

Thin verified coverage

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

Visible coverage: 13%
Verified coverage: 13%
Spread: 60.6%
Last verified: Oct 3, 2026

60%capability fit

textcodevisiondocument6 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 / text27 benchmarks60.8%

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 #169 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 19
Percentile: 69.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.

69.7% percentile inside its fair comparison set

19Raw 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 #309 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 15.7%
Percentile: 32%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.

32% percentile inside its fair comparison set

15.7%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 #234 · Source label: Qwen3.5 27B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 18.5%
Percentile: 48.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.

48.6% percentile inside its fair comparison set

18.5%Raw benchmark value

IFBench

AA · Chat / text · Objective

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

Rank #126 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 46.9%
Percentile: 64.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `ifbench`.

64.1% percentile inside its fair comparison set

46.9%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 #133 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $0.3 /1M input tokens
Percentile: 63.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.

63.8% percentile inside its fair comparison set

$0.3 /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 #177 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: $2.4 /1M output tokens
Percentile: 50.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.

50.6% percentile inside its fair comparison set

$2.4 /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 #153 · Source label: Qwen3.5 27B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 75.4 tokens/s
Percentile: 42%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

42% percentile inside its fair comparison set

75.4 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 #181 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 5.71s
Percentile: 31.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.

31.8% percentile inside its fair comparison set

5.71sRaw 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 #181 · Source label: Qwen3.5 27B (Reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 32.19s
Percentile: 31.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.

31.3% percentile inside its fair comparison set

32.19sRaw 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 #135 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,409
Percentile: 64.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: overall. Source rank: #150. Votes: 28612. Organization: alibaba. License: Apache 2.0.

64.4% percentile inside its fair comparison set

1,409Raw benchmark valueCI 1,404 - 1,413

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 #148 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,358
Percentile: 60.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: creative_writing. Source rank: #168. Votes: 4297. Organization: alibaba. License: Apache 2.0.

60.7% percentile inside its fair comparison set

1,358Raw benchmark valueCI 1,349 - 1,367

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 #119 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,428
Percentile: 68.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: english. Source rank: #133. Votes: 12729. Organization: alibaba. License: Apache 2.0.

68.6% percentile inside its fair comparison set

1,428Raw benchmark valueCI 1,422 - 1,434

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 #136 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,393
Percentile: 64.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: exclude_ties. Source rank: #151. Votes: 20882. Organization: alibaba. License: Apache 2.0.

64.1% percentile inside its fair comparison set

1,393Raw benchmark valueCI 1,387 - 1,399

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 #132 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,428
Percentile: 65.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: hard_prompts. Source rank: #147. Votes: 18070. Organization: alibaba. License: Apache 2.0.

65.2% percentile inside its fair comparison set

1,428Raw benchmark valueCI 1,422 - 1,433

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 #120 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,444
Percentile: 68.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: hard_prompts_english. Source rank: #134. Votes: 8339. Organization: alibaba. License: Apache 2.0.

68.2% percentile inside its fair comparison set

1,444Raw benchmark valueCI 1,437 - 1,451

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 #127 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,401
Percentile: 66.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: instruction_following. Source rank: #142. Votes: 9207. Organization: alibaba. License: Apache 2.0.

66.5% percentile inside its fair comparison set

1,401Raw benchmark valueCI 1,394 - 1,407

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 #120 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,422
Percentile: 66.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: longer_query. Source rank: #135. Votes: 11035. Organization: alibaba. License: Apache 2.0.

66.5% percentile inside its fair comparison set

1,422Raw benchmark valueCI 1,416 - 1,429

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 #127 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,414
Percentile: 66.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: multi_turn. Source rank: #142. Votes: 4891. Organization: alibaba. License: Apache 2.0.

66.3% percentile inside its fair comparison set

1,414Raw benchmark valueCI 1,405 - 1,423

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 #127 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,409
Percentile: 66.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: overall. Source rank: #139. Votes: 28612. Organization: alibaba. License: Apache 2.0.

66.5% percentile inside its fair comparison set

1,409Raw benchmark valueCI 1,405 - 1,413

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 #135 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 64.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: creative_writing. Source rank: #151. Votes: 4297. Organization: alibaba. License: Apache 2.0.

64.2% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,353 - 1,372

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 #117 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,425
Percentile: 69.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: english. Source rank: #126. Votes: 12729. Organization: alibaba. License: Apache 2.0.

69.1% percentile inside its fair comparison set

1,425Raw benchmark valueCI 1,419 - 1,431

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 #127 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,392
Percentile: 66.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: exclude_ties. Source rank: #139. Votes: 20882. Organization: alibaba. License: Apache 2.0.

66.5% percentile inside its fair comparison set

1,392Raw benchmark valueCI 1,386 - 1,397

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 #128 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,414
Percentile: 66.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: hard_prompts. Source rank: #140. Votes: 18070. Organization: alibaba. License: Apache 2.0.

66.2% percentile inside its fair comparison set

1,414Raw benchmark valueCI 1,409 - 1,420

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 #118 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,428
Percentile: 68.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: hard_prompts_english. Source rank: #129. Votes: 8339. Organization: alibaba. License: Apache 2.0.

68.7% percentile inside its fair comparison set

1,428Raw benchmark valueCI 1,421 - 1,435

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 #124 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,393
Percentile: 67.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: instruction_following. Source rank: #137. Votes: 9207. Organization: alibaba. License: Apache 2.0.

67.3% percentile inside its fair comparison set

1,393Raw benchmark valueCI 1,386 - 1,399

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 #113 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,413
Percentile: 68.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: longer_query. Source rank: #121. Votes: 11035. Organization: alibaba. License: Apache 2.0.

68.5% percentile inside its fair comparison set

1,413Raw benchmark valueCI 1,407 - 1,419

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 #121 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,410
Percentile: 67.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: multi_turn. Source rank: #134. Votes: 4891. Organization: alibaba. License: Apache 2.0.

67.9% percentile inside its fair comparison set

1,410Raw benchmark valueCI 1,401 - 1,419

Coding6 benchmarks52.7%

Terminal-Bench Hard

AA · Coding · Objective

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

Rank #61 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 31.8%
Percentile: 82.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `terminalbenchHard`.

82.5% percentile inside its fair comparison set

31.8%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 #137 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,452
Percentile: 63.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: coding. Source rank: #152. Votes: 7939. Organization: alibaba. License: Apache 2.0.

63.3% percentile inside its fair comparison set

1,452Raw benchmark valueCI 1,444 - 1,459

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 #124 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,427
Percentile: 66.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: coding. Source rank: #136. Votes: 7939. Organization: alibaba. License: Apache 2.0.

66.8% percentile inside its fair comparison set

1,427Raw benchmark valueCI 1,420 - 1,434

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 #99 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,357
Percentile: 22.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: overall. Source rank: #104. Votes: 7794. Organization: alibaba. License: Apache 2.0.

22.8% percentile inside its fair comparison set

1,357Raw benchmark valueCI 1,349 - 1,365

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 #90 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,386
Percentile: 28.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: webdev-html. Source rank: #95. Votes: 940. Organization: alibaba. License: Apache 2.0.

28.8% percentile inside its fair comparison set

1,386Raw benchmark valueCI 1,366 - 1,407

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 #96 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,346
Percentile: 15.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: webdev-react. Source rank: #103. Votes: 6763. Organization: alibaba. License: Apache 2.0.

15.9% percentile inside its fair comparison set

1,346Raw benchmark valueCI 1,337 - 1,354

Reasoning / math / science5 benchmarks69.1%

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 #180 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 13.9%
Percentile: 66.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `hle`.

66.2% percentile inside its fair comparison set

13.9%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 #92 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 84.2%
Percentile: 81.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `gpqa`.

81.5% percentile inside its fair comparison set

84.2%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 #189 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 0.3%
Percentile: 59.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `critpt`.

59.5% percentile inside its fair comparison set

0.3%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 #107 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,423
Percentile: 70.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: math. Source rank: #119. Votes: 1744. Organization: alibaba. License: Apache 2.0.

70.5% percentile inside its fair comparison set

1,423Raw benchmark valueCI 1,409 - 1,437

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 #92 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,429
Percentile: 74.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: math. Source rank: #100. Votes: 1744. Organization: alibaba. License: Apache 2.0.

74.7% percentile inside its fair comparison set

1,429Raw benchmark valueCI 1,415 - 1,443

Professional reasoning18 benchmarksn/a

Text Arena · Expert

AR · Professional reasoning · Human

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

Rank #116 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,441
Percentile: 64.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: expert. Source rank: #129. Votes: 2646. Organization: alibaba. License: Apache 2.0.

64.8% percentile inside its fair comparison set

1,441Raw benchmark valueCI 1,429 - 1,453

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 #132 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,408
Percentile: 64.5%
Last updated: recent
Eligibility: headline eligible

64.5% percentile inside its fair comparison set

1,408Raw benchmark valueCI 1,400 - 1,416

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 #143 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,363
Percentile: 62%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_entertainment_and_sports_and_media. Source rank: #162. Votes: 5373. Organization: alibaba. License: Apache 2.0.

62% percentile inside its fair comparison set

1,363Raw benchmark valueCI 1,354 - 1,371

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

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

Rank #140 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,406
Percentile: 60.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_legal_and_government. Source rank: #157. Votes: 2184. Organization: alibaba. License: Apache 2.0.

60.2% percentile inside its fair comparison set

1,406Raw benchmark valueCI 1,393 - 1,419

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 #111 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,441
Percentile: 70.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_life_and_physical_and_social_science. Source rank: #124. Votes: 4772. Organization: alibaba. License: Apache 2.0.

70.6% percentile inside its fair comparison set

1,441Raw benchmark valueCI 1,432 - 1,449

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

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

Rank #113 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,428
Percentile: 68.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_mathematical. Source rank: #124. Votes: 1681. Organization: alibaba. License: Apache 2.0.

68.5% percentile inside its fair comparison set

1,428Raw benchmark valueCI 1,413 - 1,443

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

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

Rank #133 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,431
Percentile: 61.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_medicine_and_healthcare. Source rank: #148. Votes: 2046. Organization: alibaba. License: Apache 2.0.

61.8% percentile inside its fair comparison set

1,431Raw benchmark valueCI 1,417 - 1,444

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 #129 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,446
Percentile: 66%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_software_and_it_services. Source rank: #143. Votes: 11345. Organization: alibaba. License: Apache 2.0.

66% percentile inside its fair comparison set

1,446Raw benchmark valueCI 1,440 - 1,453

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 #139 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,380
Percentile: 63.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_writing_and_literature_and_language. Source rank: #157. Votes: 6479. Organization: alibaba. License: Apache 2.0.

63.2% percentile inside its fair comparison set

1,380Raw benchmark valueCI 1,373 - 1,388

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

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

Rank #108 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,428
Percentile: 67.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: expert. Source rank: #116. Votes: 2646. Organization: alibaba. License: Apache 2.0.

67.3% percentile inside its fair comparison set

1,428Raw benchmark valueCI 1,416 - 1,440

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 #126 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,398
Percentile: 66.1%
Last updated: recent
Eligibility: headline eligible

66.1% percentile inside its fair comparison set

1,398Raw benchmark valueCI 1,390 - 1,407

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 #135 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,363
Percentile: 64.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_entertainment_and_sports_and_media. Source rank: #150. Votes: 5373. Organization: alibaba. License: Apache 2.0.

64.2% percentile inside its fair comparison set

1,363Raw benchmark valueCI 1,355 - 1,372

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 #135 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,405
Percentile: 61.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_legal_and_government. Source rank: #149. Votes: 2184. Organization: alibaba. License: Apache 2.0.

61.6% percentile inside its fair comparison set

1,405Raw benchmark valueCI 1,392 - 1,418

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 #103 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,436
Percentile: 72.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_life_and_physical_and_social_science. Source rank: #111. Votes: 4772. Organization: alibaba. License: Apache 2.0.

72.7% percentile inside its fair comparison set

1,436Raw benchmark valueCI 1,427 - 1,445

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

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

Rank #107 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,429
Percentile: 70.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_mathematical. Source rank: #114. Votes: 1681. Organization: alibaba. License: Apache 2.0.

70.2% percentile inside its fair comparison set

1,429Raw benchmark valueCI 1,414 - 1,443

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 #114 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,424
Percentile: 67.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_medicine_and_healthcare. Source rank: #123. Votes: 2046. Organization: alibaba. License: Apache 2.0.

67.3% percentile inside its fair comparison set

1,424Raw benchmark valueCI 1,411 - 1,437

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 #121 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,431
Percentile: 68.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_software_and_it_services. Source rank: #133. Votes: 11345. Organization: alibaba. License: Apache 2.0.

68.1% percentile inside its fair comparison set

1,431Raw benchmark valueCI 1,425 - 1,437

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 #132 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,381
Percentile: 65.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: industry_writing_and_literature_and_language. Source rank: #146. Votes: 6479. Organization: alibaba. License: Apache 2.0.

65.1% percentile inside its fair comparison set

1,381Raw benchmark valueCI 1,373 - 1,388

Search / tool use1 benchmark83.5%

Tau2-Bench Telecom

AA · Search / tool use · Objective

It matters when the model must browse, call tools, and recover useful answers from external systems.

Rank #57 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 87.1%
Percentile: 83.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `tau2`.

83.5% percentile inside its fair comparison set

87.1%Raw benchmark value

Long context1 benchmark61.3%

Long Context Reasoning

AA · Long context · Objective

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

Rank #178 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 64%
Percentile: 61.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `lcr`.

61.3% percentile inside its fair comparison set

64%Raw benchmark value

Vision understanding19 benchmarks54.4%

MMMU-Pro

AA · Vision understanding · Objective

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

Rank #95 · Source label: Qwen3.5 27B (Non-reasoning)

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Artificial Analysis
Raw value: 70%
Percentile: 54.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Artificial Analysis public leaderboard field `mmmuPro`.

54.8% percentile inside its fair comparison set

70%Raw benchmark value

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 #65 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,216
Percentile: 54%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: overall. Source rank: #73. Votes: 28993. Organization: alibaba. License: Apache 2.0.

54% percentile inside its fair comparison set

1,216Raw benchmark valueCI 1,211 - 1,222

Vision Arena · Captioning

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: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,097
Percentile: 6.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: captioning. Source rank: #32. Votes: 61. Organization: alibaba. License: Apache 2.0.

6.5% percentile inside its fair comparison set

1,097Raw benchmark valueCI 1,015 - 1,179

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 #63 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,198
Percentile: 24.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: creative_writing_vision. Source rank: #72. Votes: 2014. Organization: alibaba. License: Apache 2.0.

24.4% percentile inside its fair comparison set

1,198Raw benchmark valueCI 1,184 - 1,212

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 #66 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,239
Percentile: 34.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: diagram. Source rank: #75. Votes: 7586. Organization: alibaba. License: Apache 2.0.

34.3% percentile inside its fair comparison set

1,239Raw benchmark valueCI 1,231 - 1,247

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 #63 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,221
Percentile: 55.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: english. Source rank: #71. Votes: 11319. Organization: alibaba. License: Apache 2.0.

55.1% percentile inside its fair comparison set

1,221Raw benchmark valueCI 1,214 - 1,229

Vision Arena · Entity Recognition

AR · Vision understanding · Human

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

Rank #36 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,144
Percentile: 12.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: entity_recognition. Source rank: #45. Votes: 180. Organization: alibaba. License: Apache 2.0.

12.5% percentile inside its fair comparison set

1,144Raw benchmark valueCI 1,101 - 1,186

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 #68 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,241
Percentile: 30.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: homework. Source rank: #77. Votes: 3651. Organization: alibaba. License: Apache 2.0.

30.2% percentile inside its fair comparison set

1,241Raw benchmark valueCI 1,231 - 1,252

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 #57 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,192
Percentile: 23.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: humor. Source rank: #66. Votes: 1130. Organization: alibaba. License: Apache 2.0.

23.3% percentile inside its fair comparison set

1,192Raw benchmark valueCI 1,174 - 1,210

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 #65 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,228
Percentile: 36%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: ocr. Source rank: #74. Votes: 20452. Organization: alibaba. License: Apache 2.0.

36% percentile inside its fair comparison set

1,228Raw benchmark valueCI 1,223 - 1,234

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 #63 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,240
Percentile: 55.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: overall. Source rank: #69. Votes: 28993. Organization: alibaba. License: Apache 2.0.

55.4% percentile inside its fair comparison set

1,240Raw benchmark valueCI 1,235 - 1,246

Vision Arena · Captioning · 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 #26 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,172
Percentile: 19.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: captioning. Source rank: #27. Votes: 61. Organization: alibaba. License: Apache 2.0.

19.4% percentile inside its fair comparison set

1,172Raw benchmark valueCI 1,091 - 1,253

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 #58 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,224
Percentile: 30.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: creative_writing_vision. Source rank: #66. Votes: 2014. Organization: alibaba. License: Apache 2.0.

30.5% percentile inside its fair comparison set

1,224Raw benchmark valueCI 1,210 - 1,238

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 #65 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,248
Percentile: 35.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: diagram. Source rank: #71. Votes: 7586. Organization: alibaba. License: Apache 2.0.

35.4% percentile inside its fair comparison set

1,248Raw benchmark valueCI 1,240 - 1,256

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 #59 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,245
Percentile: 58%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: english. Source rank: #64. Votes: 11319. Organization: alibaba. License: Apache 2.0.

58% percentile inside its fair comparison set

1,245Raw benchmark valueCI 1,238 - 1,253

Vision Arena · Entity Recognition · 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 #30 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,205
Percentile: 27.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: entity_recognition. Source rank: #37. Votes: 180. Organization: alibaba. License: Apache 2.0.

27.5% percentile inside its fair comparison set

1,205Raw benchmark valueCI 1,162 - 1,247

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 #59 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,254
Percentile: 39.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: homework. Source rank: #66. Votes: 3651. Organization: alibaba. License: Apache 2.0.

39.6% percentile inside its fair comparison set

1,254Raw benchmark valueCI 1,244 - 1,265

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 #50 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,223
Percentile: 32.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: humor. Source rank: #58. Votes: 1130. Organization: alibaba. License: Apache 2.0.

32.9% percentile inside its fair comparison set

1,223Raw benchmark valueCI 1,205 - 1,242

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 #64 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,245
Percentile: 37%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: ocr. Source rank: #70. Votes: 20452. Organization: alibaba. License: Apache 2.0.

37% percentile inside its fair comparison set

1,245Raw benchmark valueCI 1,240 - 1,251

Multilingual20 benchmarksn/a

Text Arena · Chinese

AR · Multilingual · Human

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

Rank #98 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,472
Percentile: 72.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: chinese. Source rank: #105. Votes: 2041. Organization: alibaba. License: Apache 2.0.

72.4% percentile inside its fair comparison set

1,472Raw benchmark valueCI 1,459 - 1,486

Text Arena · French

AR · Multilingual · Human

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

Rank #130 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,416
Percentile: 49.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: french. Source rank: #146. Votes: 849. Organization: alibaba. License: Apache 2.0.

49.8% percentile inside its fair comparison set

1,416Raw benchmark valueCI 1,394 - 1,437

Text Arena · German

AR · Multilingual · Human

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

Rank #116 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,391
Percentile: 57.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: german. Source rank: #132. Votes: 452. Organization: alibaba. License: Apache 2.0.

57.7% percentile inside its fair comparison set

1,391Raw benchmark valueCI 1,363 - 1,418

Text Arena · Japanese

AR · Multilingual · Human

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

Rank #117 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,338
Percentile: 52.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: japanese. Source rank: #135. Votes: 224. Organization: alibaba. License: Apache 2.0.

52.3% percentile inside its fair comparison set

1,338Raw benchmark valueCI 1,298 - 1,378

Text Arena · Korean

AR · Multilingual · Human

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

Rank #117 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,350
Percentile: 52.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: korean. Source rank: #133. Votes: 446. Organization: alibaba. License: Apache 2.0.

52.5% percentile inside its fair comparison set

1,350Raw benchmark valueCI 1,322 - 1,379

Text Arena · Non English

AR · Multilingual · Human

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

Rank #140 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,387
Percentile: 63%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: non_english. Source rank: #157. Votes: 15883. Organization: alibaba. License: Apache 2.0.

63% percentile inside its fair comparison set

1,387Raw benchmark valueCI 1,381 - 1,392

Text Arena · Polish

AR · Multilingual · Human

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

Rank #136 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,366
Percentile: 32.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: polish. Source rank: #159. Votes: 538. Organization: alibaba. License: Apache 2.0.

32.2% percentile inside its fair comparison set

1,366Raw benchmark valueCI 1,341 - 1,392

Text Arena · Russian

AR · Multilingual · Human

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

Rank #142 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,390
Percentile: 58.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: russian. Source rank: #160. Votes: 3314. Organization: alibaba. License: Apache 2.0.

58.7% percentile inside its fair comparison set

1,390Raw benchmark valueCI 1,379 - 1,400

Text Arena · Spanish

AR · Multilingual · Human

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

Rank #130 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,398
Percentile: 50.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: spanish. Source rank: #147. Votes: 940. Organization: alibaba. License: Apache 2.0.

50.2% percentile inside its fair comparison set

1,398Raw benchmark valueCI 1,379 - 1,418

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #72 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,478
Percentile: 79.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: chinese. Source rank: #75. Votes: 2041. Organization: alibaba. License: Apache 2.0.

79.8% percentile inside its fair comparison set

1,478Raw benchmark valueCI 1,465 - 1,491

Text Arena · French · No Style Control

AR · Multilingual · Human

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

Rank #123 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,409
Percentile: 52.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: french. Source rank: #136. Votes: 849. Organization: alibaba. License: Apache 2.0.

52.5% percentile inside its fair comparison set

1,409Raw benchmark valueCI 1,388 - 1,430

Text Arena · German · No Style Control

AR · Multilingual · Human

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

Rank #110 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,393
Percentile: 59.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: german. Source rank: #124. Votes: 452. Organization: alibaba. License: Apache 2.0.

59.9% percentile inside its fair comparison set

1,393Raw benchmark valueCI 1,366 - 1,420

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

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

Rank #108 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,345
Percentile: 56%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: japanese. Source rank: #123. Votes: 224. Organization: alibaba. License: Apache 2.0.

56% percentile inside its fair comparison set

1,345Raw benchmark valueCI 1,304 - 1,385

Text Arena · Korean · No Style Control

AR · Multilingual · Human

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

Rank #108 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,358
Percentile: 56.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: korean. Source rank: #122. Votes: 446. Organization: alibaba. License: Apache 2.0.

56.1% percentile inside its fair comparison set

1,358Raw benchmark valueCI 1,329 - 1,387

Text Arena · Non English · No Style Control

AR · Multilingual · Human

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

Rank #133 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,390
Percentile: 64.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: non_english. Source rank: #147. Votes: 15883. Organization: alibaba. License: Apache 2.0.

64.9% percentile inside its fair comparison set

1,390Raw benchmark valueCI 1,384 - 1,395

Text Arena · Polish · No Style Control

AR · Multilingual · Human

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

Rank #123 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,375
Percentile: 38.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: polish. Source rank: #141. Votes: 538. Organization: alibaba. License: Apache 2.0.

38.7% percentile inside its fair comparison set

1,375Raw benchmark valueCI 1,351 - 1,400

Text Arena · Russian · No Style Control

AR · Multilingual · Human

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

Rank #134 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,390
Percentile: 61%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: russian. Source rank: #149. Votes: 3314. Organization: alibaba. License: Apache 2.0.

61% percentile inside its fair comparison set

1,390Raw benchmark valueCI 1,379 - 1,400

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

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

Rank #118 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,407
Percentile: 54.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: spanish. Source rank: #129. Votes: 940. Organization: alibaba. License: Apache 2.0.

54.8% percentile inside its fair comparison set

1,407Raw benchmark valueCI 1,387 - 1,426

Vision Arena · Chinese

AR · Multilingual · Human

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

Rank #52 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,271
Percentile: 51%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: chinese. Source rank: #58. Votes: 1923. Organization: alibaba. License: Apache 2.0.

51% percentile inside its fair comparison set

1,271Raw benchmark valueCI 1,255 - 1,287

Vision Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #45 · Source label: qwen3.5-27b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,301
Percentile: 57.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3.5-27b`. Category: chinese. Source rank: #49. Votes: 1923. Organization: alibaba. License: Apache 2.0.

57.7% percentile inside its fair comparison set

1,301Raw benchmark valueCI 1,285 - 1,317

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

Arena

Oct 3, 2026

source →

official

Artificial Analysis

Oct 3, 2026

source →

Qwen3.5 27B

Current snapshot.

Source-linked scores by benchmark

Source links and registry checks

Loading model evidence.

Qwen3.5 27B

Current snapshot.

Source-linked scores by benchmark

Source links and registry checks