Model profile · Qwen

Qwen3 235B A22B

Open weightsmid · registry tag 2026 open generalist

Thin verified coverage

Reads as thin verified coverage across the resolved source data.

Visible coverage: 7.5%
Verified coverage: 7.5%
Spread: 62.8%
Last verified: Jun 20, 2026

33%bench fit

textcodevision3 aliases28 official source links

Open compare

Data version

Current snapshot.

Data version Jun 20, 2026Model list checked9 providers · 1081 tracked modelsPage refreshed Jul 5, 2026

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

Source-linked scores by benchmark

Each row keeps the benchmark source, source type, raw metric, and percentile inside its fair comparison set.

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

Chat / text18 benchmarks61.3%

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 #122

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,375
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: overall. Source rank: #147. Votes: 26267. Organization: alibaba. License: Apache 2.0.

62.8% percentile inside its fair comparison set

1,375Raw benchmark valueCI 1,370 - 1,380

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 #137

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,323
Percentile: 57.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: creative_writing. Source rank: #167. Votes: 3579. Organization: alibaba. License: Apache 2.0.

57.9% percentile inside its fair comparison set

1,323Raw benchmark valueCI 1,313 - 1,333

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 #122

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,387
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: english. Source rank: #147. Votes: 13453. Organization: alibaba. License: Apache 2.0.

62.8% percentile inside its fair comparison set

1,387Raw benchmark valueCI 1,382 - 1,393

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,346
Percentile: 62.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: exclude_ties. Source rank: #150. Votes: 18072. Organization: alibaba. License: Apache 2.0.

62.2% percentile inside its fair comparison set

1,346Raw benchmark valueCI 1,339 - 1,352

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 #122

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,392
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: hard_prompts. Source rank: #148. Votes: 10495. Organization: alibaba. License: Apache 2.0.

62.8% percentile inside its fair comparison set

1,392Raw benchmark valueCI 1,385 - 1,398

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,403
Percentile: 62%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: hard_prompts_english. Source rank: #150. Votes: 5751. Organization: alibaba. License: Apache 2.0.

62% percentile inside its fair comparison set

1,403Raw benchmark valueCI 1,395 - 1,411

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 #129

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,357
Percentile: 60.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: instruction_following. Source rank: #156. Votes: 6192. Organization: alibaba. License: Apache 2.0.

60.6% percentile inside its fair comparison set

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

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 #123

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,378
Percentile: 59.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: longer_query. Source rank: #149. Votes: 4584. Organization: alibaba. License: Apache 2.0.

59.9% percentile inside its fair comparison set

1,378Raw benchmark valueCI 1,369 - 1,387

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,372
Percentile: 61.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: multi_turn. Source rank: #150. Votes: 4281. Organization: alibaba. License: Apache 2.0.

61.9% percentile inside its fair comparison set

1,372Raw benchmark valueCI 1,363 - 1,381

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 #121

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,366
Percentile: 63.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: overall. Source rank: #145. Votes: 26267. Organization: alibaba. License: Apache 2.0.

63.1% percentile inside its fair comparison set

1,366Raw benchmark valueCI 1,361 - 1,370

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

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,316
Percentile: 58.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: creative_writing. Source rank: #163. Votes: 3579. Organization: alibaba. License: Apache 2.0.

58.5% percentile inside its fair comparison set

1,316Raw benchmark valueCI 1,306 - 1,326

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 #126

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,377
Percentile: 61.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: english. Source rank: #149. Votes: 13453. Organization: alibaba. License: Apache 2.0.

61.5% percentile inside its fair comparison set

1,377Raw benchmark valueCI 1,371 - 1,382

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 #120

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,332
Percentile: 63.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: exclude_ties. Source rank: #144. Votes: 18072. Organization: alibaba. License: Apache 2.0.

63.4% percentile inside its fair comparison set

1,332Raw benchmark valueCI 1,325 - 1,338

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 #126

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 61.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: hard_prompts. Source rank: #152. Votes: 10495. Organization: alibaba. License: Apache 2.0.

61.5% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,356 - 1,369

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 #133

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,372
Percentile: 59.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: hard_prompts_english. Source rank: #159. Votes: 5751. Organization: alibaba. License: Apache 2.0.

59.3% percentile inside its fair comparison set

1,372Raw benchmark valueCI 1,364 - 1,380

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 #130

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,333
Percentile: 60.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: instruction_following. Source rank: #157. Votes: 6192. Organization: alibaba. License: Apache 2.0.

60.3% percentile inside its fair comparison set

1,333Raw benchmark valueCI 1,325 - 1,341

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,352
Percentile: 59.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: longer_query. Source rank: #152. Votes: 4584. Organization: alibaba. License: Apache 2.0.

59.5% percentile inside its fair comparison set

1,352Raw benchmark valueCI 1,343 - 1,361

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

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: multi_turn. Source rank: #146. Votes: 4281. Organization: alibaba. License: Apache 2.0.

62.8% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,353 - 1,371

Coding4 benchmarks41.7%

LiveCodeBench

VALS-AI · Coding · Objective

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

Rank #57 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 70.6%
Percentile: 37.8%
Last updated: recent
Eligibility: headline eligible

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

37.8% percentile inside its fair comparison set

70.6%Raw benchmark valueCI 68.4% - 72.9%

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 #111

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,434
Percentile: 65.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: coding. Source rank: #133. Votes: 4340. Organization: alibaba. License: Apache 2.0.

65.6% percentile inside its fair comparison set

1,434Raw benchmark valueCI 1,425 - 1,443

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 #118

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,386
Percentile: 63.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: coding. Source rank: #142. Votes: 4340. Organization: alibaba. License: Apache 2.0.

63.4% percentile inside its fair comparison set

1,386Raw benchmark valueCI 1,377 - 1,395

IOI

VALS-AI · Coding · Objective

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

Rank #45 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

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

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

0% percentile inside its fair comparison set

0%Raw benchmark valueCI 0% - 0%

Reasoning / math / science5 benchmarks42.7%

GPQA Diamond

VALS-AI · Reasoning / math / science · Objective

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

Rank #64 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 70.2%
Percentile: 30.3%
Last updated: recent
Eligibility: headline eligible

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

30.3% percentile inside its fair comparison set

70.2%Raw benchmark valueCI 65.6% - 74.9%

MMLU Pro

VALS-AI · Reasoning / math / science · Objective

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

Rank #52 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 81.2%
Percentile: 42.7%
Last updated: recent
Eligibility: headline eligible

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

42.7% percentile inside its fair comparison set

81.2%Raw benchmark valueCI 80.5% - 82%

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 #108

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,393
Percentile: 65.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: math. Source rank: #131. Votes: 1604. Organization: alibaba. License: Apache 2.0.

65.9% percentile inside its fair comparison set

1,393Raw benchmark valueCI 1,379 - 1,407

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 #102

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,395
Percentile: 67.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: math. Source rank: #122. Votes: 1604. Organization: alibaba. License: Apache 2.0.

67.8% percentile inside its fair comparison set

1,395Raw benchmark valueCI 1,382 - 1,409

EnigmaEval

SL · Reasoning / math / science · Rubric

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

Rank #15

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 57%
Percentile: 6.7%
Last updated: aging
Eligibility: headline eligible

6.7% percentile inside its fair comparison set

57%Raw benchmark value

Professional reasoning20 benchmarks57.7%

LegalBench

VALS-AI · Professional reasoning · Objective

Academic legal reasoning tasks.

Rank #56 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 80.2%
Percentile: 38.9%
Last updated: recent
Eligibility: headline eligible

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

38.9% percentile inside its fair comparison set

80.2%Raw benchmark valueCI 79.3% - 81%

TaxEval v2

VALS-AI · Professional reasoning · Objective

Answer quality on tax questions and responses.

Rank #58 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 70.6%
Percentile: 37.4%
Last updated: recent
Eligibility: headline eligible

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

37.4% percentile inside its fair comparison set

70.6%Raw benchmark valueCI 68.9% - 72.4%

Text Arena · Expert

AR · Professional reasoning · Human

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

Rank #122

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,385
Percentile: 56%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: expert. Source rank: #148. Votes: 1319. Organization: alibaba. License: Apache 2.0.

56% percentile inside its fair comparison set

1,385Raw benchmark valueCI 1,369 - 1,401

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 #125

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,369
Percentile: 61%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_business_and_management_and_financial_operations. Source rank: #150. Votes: 3964. Organization: alibaba. License: Apache 2.0.

61% percentile inside its fair comparison set

1,369Raw benchmark valueCI 1,360 - 1,379

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 #135

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,324
Percentile: 58.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_entertainment_and_sports_and_media. Source rank: #164. Votes: 4743. Organization: alibaba. License: Apache 2.0.

58.5% percentile inside its fair comparison set

1,324Raw benchmark valueCI 1,316 - 1,333

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

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

Rank #146

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,364
Percentile: 51.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_legal_and_government. Source rank: #174. Votes: 1735. Organization: alibaba. License: Apache 2.0.

51.3% percentile inside its fair comparison set

1,364Raw benchmark valueCI 1,351 - 1,378

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 #130

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,383
Percentile: 60.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_life_and_physical_and_social_science. Source rank: #157. Votes: 4590. Organization: alibaba. License: Apache 2.0.

60.1% percentile inside its fair comparison set

1,383Raw benchmark valueCI 1,374 - 1,392

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

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

Rank #102

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,401
Percentile: 67.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_mathematical. Source rank: #125. Votes: 1501. Organization: alibaba. License: Apache 2.0.

67.2% percentile inside its fair comparison set

1,401Raw benchmark valueCI 1,386 - 1,416

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

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,381
Percentile: 55.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_medicine_and_healthcare. Source rank: #159. Votes: 1549. Organization: alibaba. License: Apache 2.0.

55.3% percentile inside its fair comparison set

1,381Raw benchmark valueCI 1,366 - 1,397

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 #118

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,417
Percentile: 64%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_software_and_it_services. Source rank: #142. Votes: 8098. Organization: alibaba. License: Apache 2.0.

64% percentile inside its fair comparison set

1,417Raw benchmark valueCI 1,410 - 1,424

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 #129

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,345
Percentile: 60.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_writing_and_literature_and_language. Source rank: #156. Votes: 6170. Organization: alibaba. License: Apache 2.0.

60.5% percentile inside its fair comparison set

1,345Raw benchmark valueCI 1,337 - 1,353

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

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

Rank #125

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,348
Percentile: 54.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: expert. Source rank: #151. Votes: 1319. Organization: alibaba. License: Apache 2.0.

54.9% percentile inside its fair comparison set

1,348Raw benchmark valueCI 1,332 - 1,364

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,349
Percentile: 61.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_business_and_management_and_financial_operations. Source rank: #148. Votes: 3964. Organization: alibaba. License: Apache 2.0.

61.3% percentile inside its fair comparison set

1,349Raw benchmark valueCI 1,339 - 1,358

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 #129

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,316
Percentile: 60.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_entertainment_and_sports_and_media. Source rank: #156. Votes: 4743. Organization: alibaba. License: Apache 2.0.

60.4% percentile inside its fair comparison set

1,316Raw benchmark valueCI 1,308 - 1,325

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 #130

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,355
Percentile: 56.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_legal_and_government. Source rank: #156. Votes: 1735. Organization: alibaba. License: Apache 2.0.

56.7% percentile inside its fair comparison set

1,355Raw benchmark valueCI 1,341 - 1,369

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,372
Percentile: 61.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_life_and_physical_and_social_science. Source rank: #148. Votes: 4590. Organization: alibaba. License: Apache 2.0.

61.9% percentile inside its fair comparison set

1,372Raw benchmark valueCI 1,363 - 1,381

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

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

Rank #103

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,397
Percentile: 66.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_mathematical. Source rank: #123. Votes: 1501. Organization: alibaba. License: Apache 2.0.

66.9% percentile inside its fair comparison set

1,397Raw benchmark valueCI 1,382 - 1,412

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 #127

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,363
Percentile: 57.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_medicine_and_healthcare. Source rank: #152. Votes: 1549. Organization: alibaba. License: Apache 2.0.

57.3% percentile inside its fair comparison set

1,363Raw benchmark valueCI 1,348 - 1,378

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 #118

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,387
Percentile: 64%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_software_and_it_services. Source rank: #141. Votes: 8098. Organization: alibaba. License: Apache 2.0.

64% percentile inside its fair comparison set

1,387Raw benchmark valueCI 1,380 - 1,394

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 #126

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,336
Percentile: 61.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_writing_and_literature_and_language. Source rank: #153. Votes: 6170. Organization: alibaba. License: Apache 2.0.

61.4% percentile inside its fair comparison set

1,336Raw benchmark valueCI 1,329 - 1,344

Vision understanding1 benchmark7.1%

VISTA

SL · Vision understanding · Rubric

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

Rank #14

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 71%
Percentile: 7.1%
Last updated: aging
Eligibility: headline eligible

7.1% percentile inside its fair comparison set

71%Raw benchmark value

Document understanding1 benchmark14.3%

Multimodal mix

OC · Document understanding · Objective

It matters when the job is reading PDFs, tables, forms, or mixed-layout documents rather than plain chat.

Rank #13 · Source label: qwen3-235b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: OpenCompass
Raw value: 70.8%
Percentile: 14.3%
Last updated: recent
Eligibility: headline eligible

14.3% percentile inside its fair comparison set

70.8%Raw benchmark value

Embeddings / retrieval1 benchmark50%

Retrieval

MTEB · Embeddings / retrieval · Retrieval

It is one of the few direct signals for retrieval stacks, where embedding quality matters more than chat style.

Rank #6 · Source label: qwen3-235b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: MTEB
Raw value: 56.2 ndcg
Percentile: 50%
Last updated: aging
Eligibility: headline eligible

Strong open retrieval score.

50% percentile inside its fair comparison set

56.2 ndcgRaw benchmark value

Multilingual14 benchmarks54%

Text Arena · Chinese

AR · Multilingual · Human

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

Rank #122

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,394
Percentile: 59%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: chinese. Source rank: #148. Votes: 1254. Organization: alibaba. License: Apache 2.0.

59% percentile inside its fair comparison set

1,394Raw benchmark valueCI 1,377 - 1,410

Text Arena · French

AR · Multilingual · Human

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

Rank #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,374
Percentile: 43.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: french. Source rank: #147. Votes: 326. Organization: alibaba. License: Apache 2.0.

43.1% percentile inside its fair comparison set

1,374Raw benchmark valueCI 1,342 - 1,407

Text Arena · German

AR · Multilingual · Human

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

Rank #91

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,382
Percentile: 62%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: german. Source rank: #111. Votes: 727. Organization: alibaba. License: Apache 2.0.

62% percentile inside its fair comparison set

1,382Raw benchmark valueCI 1,360 - 1,404

Text Arena · Japanese

AR · Multilingual · Human

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

Rank #100

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,309
Percentile: 51.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: japanese. Source rank: #123. Votes: 646. Organization: alibaba. License: Apache 2.0.

51.2% percentile inside its fair comparison set

1,309Raw benchmark valueCI 1,285 - 1,333

Text Arena · Korean

AR · Multilingual · Human

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

Rank #110

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,318
Percentile: 47.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: korean. Source rank: #134. Votes: 471. Organization: alibaba. License: Apache 2.0.

47.6% percentile inside its fair comparison set

1,318Raw benchmark valueCI 1,290 - 1,346

Text Arena · Russian

AR · Multilingual · Human

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

Rank #129

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,354
Percentile: 55.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: russian. Source rank: #156. Votes: 1857. Organization: alibaba. License: Apache 2.0.

55.7% percentile inside its fair comparison set

1,354Raw benchmark valueCI 1,341 - 1,367

Text Arena · Spanish

AR · Multilingual · Human

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

Rank #112

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,369
Percentile: 48.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: spanish. Source rank: #137. Votes: 421. Organization: alibaba. License: Apache 2.0.

48.1% percentile inside its fair comparison set

1,369Raw benchmark valueCI 1,341 - 1,398

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #119

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,380
Percentile: 60%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: chinese. Source rank: #144. Votes: 1254. Organization: alibaba. License: Apache 2.0.

60% percentile inside its fair comparison set

1,380Raw benchmark valueCI 1,363 - 1,396

Text Arena · French · No Style Control

AR · Multilingual · Human

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

Rank #113

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,368
Percentile: 48.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: french. Source rank: #134. Votes: 326. Organization: alibaba. License: Apache 2.0.

48.1% percentile inside its fair comparison set

1,368Raw benchmark valueCI 1,336 - 1,401

Text Arena · German · No Style Control

AR · Multilingual · Human

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

Rank #83

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,386
Percentile: 65.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: german. Source rank: #100. Votes: 727. Organization: alibaba. License: Apache 2.0.

65.4% percentile inside its fair comparison set

1,386Raw benchmark valueCI 1,364 - 1,408

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

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

Rank #90

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,308
Percentile: 56.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: japanese. Source rank: #111. Votes: 646. Organization: alibaba. License: Apache 2.0.

56.2% percentile inside its fair comparison set

1,308Raw benchmark valueCI 1,284 - 1,332

Text Arena · Korean · No Style Control

AR · Multilingual · Human

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

Rank #100

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,314
Percentile: 52.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: korean. Source rank: #120. Votes: 471. Organization: alibaba. License: Apache 2.0.

52.4% percentile inside its fair comparison set

1,314Raw benchmark valueCI 1,286 - 1,342

Text Arena · Russian · No Style Control

AR · Multilingual · Human

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

Rank #126

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,338
Percentile: 56.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: russian. Source rank: #153. Votes: 1857. Organization: alibaba. License: Apache 2.0.

56.7% percentile inside its fair comparison set

1,338Raw benchmark valueCI 1,325 - 1,351

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

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

Rank #108

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 50%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: spanish. Source rank: #130. Votes: 421. Organization: alibaba. License: Apache 2.0.

50% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,334 - 1,391

Source links and registry checks

official

Qwen3 announcement

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

official

Arena

Jun 20, 2026

source →

Model profile · Qwen

Qwen3 235B A22B

Open weightsmid · registry tag 2026 open generalist

Thin verified coverage

Reads as thin verified coverage across the resolved source data.

Visible coverage: 7.5%
Verified coverage: 7.5%
Spread: 62.8%
Last verified: Jun 20, 2026

33%bench fit

textcodevision3 aliases28 official source links

Open compare

Data version

Current snapshot.

Data version Jun 20, 2026Model list checked9 providers · 1081 tracked modelsPage refreshed Jul 5, 2026

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

Source-linked scores by benchmark

Each row keeps the benchmark source, source type, raw metric, and percentile inside its fair comparison set.

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

Chat / text18 benchmarks61.3%

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 #122

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,375
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: overall. Source rank: #147. Votes: 26267. Organization: alibaba. License: Apache 2.0.

62.8% percentile inside its fair comparison set

1,375Raw benchmark valueCI 1,370 - 1,380

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 #137

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,323
Percentile: 57.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: creative_writing. Source rank: #167. Votes: 3579. Organization: alibaba. License: Apache 2.0.

57.9% percentile inside its fair comparison set

1,323Raw benchmark valueCI 1,313 - 1,333

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 #122

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,387
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: english. Source rank: #147. Votes: 13453. Organization: alibaba. License: Apache 2.0.

62.8% percentile inside its fair comparison set

1,387Raw benchmark valueCI 1,382 - 1,393

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,346
Percentile: 62.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: exclude_ties. Source rank: #150. Votes: 18072. Organization: alibaba. License: Apache 2.0.

62.2% percentile inside its fair comparison set

1,346Raw benchmark valueCI 1,339 - 1,352

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 #122

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,392
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: hard_prompts. Source rank: #148. Votes: 10495. Organization: alibaba. License: Apache 2.0.

62.8% percentile inside its fair comparison set

1,392Raw benchmark valueCI 1,385 - 1,398

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,403
Percentile: 62%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: hard_prompts_english. Source rank: #150. Votes: 5751. Organization: alibaba. License: Apache 2.0.

62% percentile inside its fair comparison set

1,403Raw benchmark valueCI 1,395 - 1,411

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 #129

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,357
Percentile: 60.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: instruction_following. Source rank: #156. Votes: 6192. Organization: alibaba. License: Apache 2.0.

60.6% percentile inside its fair comparison set

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

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 #123

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,378
Percentile: 59.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: longer_query. Source rank: #149. Votes: 4584. Organization: alibaba. License: Apache 2.0.

59.9% percentile inside its fair comparison set

1,378Raw benchmark valueCI 1,369 - 1,387

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,372
Percentile: 61.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: multi_turn. Source rank: #150. Votes: 4281. Organization: alibaba. License: Apache 2.0.

61.9% percentile inside its fair comparison set

1,372Raw benchmark valueCI 1,363 - 1,381

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 #121

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,366
Percentile: 63.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: overall. Source rank: #145. Votes: 26267. Organization: alibaba. License: Apache 2.0.

63.1% percentile inside its fair comparison set

1,366Raw benchmark valueCI 1,361 - 1,370

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

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,316
Percentile: 58.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: creative_writing. Source rank: #163. Votes: 3579. Organization: alibaba. License: Apache 2.0.

58.5% percentile inside its fair comparison set

1,316Raw benchmark valueCI 1,306 - 1,326

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 #126

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,377
Percentile: 61.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: english. Source rank: #149. Votes: 13453. Organization: alibaba. License: Apache 2.0.

61.5% percentile inside its fair comparison set

1,377Raw benchmark valueCI 1,371 - 1,382

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 #120

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,332
Percentile: 63.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: exclude_ties. Source rank: #144. Votes: 18072. Organization: alibaba. License: Apache 2.0.

63.4% percentile inside its fair comparison set

1,332Raw benchmark valueCI 1,325 - 1,338

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 #126

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 61.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: hard_prompts. Source rank: #152. Votes: 10495. Organization: alibaba. License: Apache 2.0.

61.5% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,356 - 1,369

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 #133

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,372
Percentile: 59.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: hard_prompts_english. Source rank: #159. Votes: 5751. Organization: alibaba. License: Apache 2.0.

59.3% percentile inside its fair comparison set

1,372Raw benchmark valueCI 1,364 - 1,380

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 #130

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,333
Percentile: 60.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: instruction_following. Source rank: #157. Votes: 6192. Organization: alibaba. License: Apache 2.0.

60.3% percentile inside its fair comparison set

1,333Raw benchmark valueCI 1,325 - 1,341

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,352
Percentile: 59.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: longer_query. Source rank: #152. Votes: 4584. Organization: alibaba. License: Apache 2.0.

59.5% percentile inside its fair comparison set

1,352Raw benchmark valueCI 1,343 - 1,361

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

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 62.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: multi_turn. Source rank: #146. Votes: 4281. Organization: alibaba. License: Apache 2.0.

62.8% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,353 - 1,371

Coding4 benchmarks41.7%

LiveCodeBench

VALS-AI · Coding · Objective

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

Rank #57 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 70.6%
Percentile: 37.8%
Last updated: recent
Eligibility: headline eligible

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

37.8% percentile inside its fair comparison set

70.6%Raw benchmark valueCI 68.4% - 72.9%

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 #111

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,434
Percentile: 65.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: coding. Source rank: #133. Votes: 4340. Organization: alibaba. License: Apache 2.0.

65.6% percentile inside its fair comparison set

1,434Raw benchmark valueCI 1,425 - 1,443

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 #118

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,386
Percentile: 63.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: coding. Source rank: #142. Votes: 4340. Organization: alibaba. License: Apache 2.0.

63.4% percentile inside its fair comparison set

1,386Raw benchmark valueCI 1,377 - 1,395

IOI

VALS-AI · Coding · Objective

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

Rank #45 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

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

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

0% percentile inside its fair comparison set

0%Raw benchmark valueCI 0% - 0%

Reasoning / math / science5 benchmarks42.7%

GPQA Diamond

VALS-AI · Reasoning / math / science · Objective

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

Rank #64 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 70.2%
Percentile: 30.3%
Last updated: recent
Eligibility: headline eligible

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

30.3% percentile inside its fair comparison set

70.2%Raw benchmark valueCI 65.6% - 74.9%

MMLU Pro

VALS-AI · Reasoning / math / science · Objective

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

Rank #52 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 81.2%
Percentile: 42.7%
Last updated: recent
Eligibility: headline eligible

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

42.7% percentile inside its fair comparison set

81.2%Raw benchmark valueCI 80.5% - 82%

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 #108

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,393
Percentile: 65.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: math. Source rank: #131. Votes: 1604. Organization: alibaba. License: Apache 2.0.

65.9% percentile inside its fair comparison set

1,393Raw benchmark valueCI 1,379 - 1,407

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 #102

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,395
Percentile: 67.8%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: math. Source rank: #122. Votes: 1604. Organization: alibaba. License: Apache 2.0.

67.8% percentile inside its fair comparison set

1,395Raw benchmark valueCI 1,382 - 1,409

EnigmaEval

SL · Reasoning / math / science · Rubric

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

Rank #15

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 57%
Percentile: 6.7%
Last updated: aging
Eligibility: headline eligible

6.7% percentile inside its fair comparison set

57%Raw benchmark value

Professional reasoning20 benchmarks57.7%

LegalBench

VALS-AI · Professional reasoning · Objective

Academic legal reasoning tasks.

Rank #56 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 80.2%
Percentile: 38.9%
Last updated: recent
Eligibility: headline eligible

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

38.9% percentile inside its fair comparison set

80.2%Raw benchmark valueCI 79.3% - 81%

TaxEval v2

VALS-AI · Professional reasoning · Objective

Answer quality on tax questions and responses.

Rank #58 · Source label: fireworks/qwen3-235b-a22b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Vals AI
Raw value: 70.6%
Percentile: 37.4%
Last updated: recent
Eligibility: headline eligible

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

37.4% percentile inside its fair comparison set

70.6%Raw benchmark valueCI 68.9% - 72.4%

Text Arena · Expert

AR · Professional reasoning · Human

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

Rank #122

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,385
Percentile: 56%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: expert. Source rank: #148. Votes: 1319. Organization: alibaba. License: Apache 2.0.

56% percentile inside its fair comparison set

1,385Raw benchmark valueCI 1,369 - 1,401

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 #125

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,369
Percentile: 61%
Last updated: recent
Eligibility: headline eligible

61% percentile inside its fair comparison set

1,369Raw benchmark valueCI 1,360 - 1,379

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 #135

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,324
Percentile: 58.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_entertainment_and_sports_and_media. Source rank: #164. Votes: 4743. Organization: alibaba. License: Apache 2.0.

58.5% percentile inside its fair comparison set

1,324Raw benchmark valueCI 1,316 - 1,333

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

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

Rank #146

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,364
Percentile: 51.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_legal_and_government. Source rank: #174. Votes: 1735. Organization: alibaba. License: Apache 2.0.

51.3% percentile inside its fair comparison set

1,364Raw benchmark valueCI 1,351 - 1,378

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 #130

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,383
Percentile: 60.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_life_and_physical_and_social_science. Source rank: #157. Votes: 4590. Organization: alibaba. License: Apache 2.0.

60.1% percentile inside its fair comparison set

1,383Raw benchmark valueCI 1,374 - 1,392

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

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

Rank #102

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,401
Percentile: 67.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_mathematical. Source rank: #125. Votes: 1501. Organization: alibaba. License: Apache 2.0.

67.2% percentile inside its fair comparison set

1,401Raw benchmark valueCI 1,386 - 1,416

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

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,381
Percentile: 55.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_medicine_and_healthcare. Source rank: #159. Votes: 1549. Organization: alibaba. License: Apache 2.0.

55.3% percentile inside its fair comparison set

1,381Raw benchmark valueCI 1,366 - 1,397

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 #118

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,417
Percentile: 64%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_software_and_it_services. Source rank: #142. Votes: 8098. Organization: alibaba. License: Apache 2.0.

64% percentile inside its fair comparison set

1,417Raw benchmark valueCI 1,410 - 1,424

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 #129

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,345
Percentile: 60.5%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_writing_and_literature_and_language. Source rank: #156. Votes: 6170. Organization: alibaba. License: Apache 2.0.

60.5% percentile inside its fair comparison set

1,345Raw benchmark valueCI 1,337 - 1,353

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

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

Rank #125

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,348
Percentile: 54.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: expert. Source rank: #151. Votes: 1319. Organization: alibaba. License: Apache 2.0.

54.9% percentile inside its fair comparison set

1,348Raw benchmark valueCI 1,332 - 1,364

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,349
Percentile: 61.3%
Last updated: recent
Eligibility: headline eligible

61.3% percentile inside its fair comparison set

1,349Raw benchmark valueCI 1,339 - 1,358

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 #129

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,316
Percentile: 60.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_entertainment_and_sports_and_media. Source rank: #156. Votes: 4743. Organization: alibaba. License: Apache 2.0.

60.4% percentile inside its fair comparison set

1,316Raw benchmark valueCI 1,308 - 1,325

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 #130

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,355
Percentile: 56.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_legal_and_government. Source rank: #156. Votes: 1735. Organization: alibaba. License: Apache 2.0.

56.7% percentile inside its fair comparison set

1,355Raw benchmark valueCI 1,341 - 1,369

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 #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,372
Percentile: 61.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_life_and_physical_and_social_science. Source rank: #148. Votes: 4590. Organization: alibaba. License: Apache 2.0.

61.9% percentile inside its fair comparison set

1,372Raw benchmark valueCI 1,363 - 1,381

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

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

Rank #103

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,397
Percentile: 66.9%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_mathematical. Source rank: #123. Votes: 1501. Organization: alibaba. License: Apache 2.0.

66.9% percentile inside its fair comparison set

1,397Raw benchmark valueCI 1,382 - 1,412

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 #127

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,363
Percentile: 57.3%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_medicine_and_healthcare. Source rank: #152. Votes: 1549. Organization: alibaba. License: Apache 2.0.

57.3% percentile inside its fair comparison set

1,363Raw benchmark valueCI 1,348 - 1,378

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 #118

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,387
Percentile: 64%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_software_and_it_services. Source rank: #141. Votes: 8098. Organization: alibaba. License: Apache 2.0.

64% percentile inside its fair comparison set

1,387Raw benchmark valueCI 1,380 - 1,394

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 #126

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,336
Percentile: 61.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: industry_writing_and_literature_and_language. Source rank: #153. Votes: 6170. Organization: alibaba. License: Apache 2.0.

61.4% percentile inside its fair comparison set

1,336Raw benchmark valueCI 1,329 - 1,344

Vision understanding1 benchmark7.1%

VISTA

SL · Vision understanding · Rubric

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

Rank #14

verified runtimeexact direct

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Scale Labs
Raw value: 71%
Percentile: 7.1%
Last updated: aging
Eligibility: headline eligible

7.1% percentile inside its fair comparison set

71%Raw benchmark value

Document understanding1 benchmark14.3%

Multimodal mix

OC · Document understanding · Objective

It matters when the job is reading PDFs, tables, forms, or mixed-layout documents rather than plain chat.

Rank #13 · Source label: qwen3-235b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: OpenCompass
Raw value: 70.8%
Percentile: 14.3%
Last updated: recent
Eligibility: headline eligible

14.3% percentile inside its fair comparison set

70.8%Raw benchmark value

Embeddings / retrieval1 benchmark50%

Retrieval

MTEB · Embeddings / retrieval · Retrieval

It is one of the few direct signals for retrieval stacks, where embedding quality matters more than chat style.

Rank #6 · Source label: qwen3-235b

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: MTEB
Raw value: 56.2 ndcg
Percentile: 50%
Last updated: aging
Eligibility: headline eligible

Strong open retrieval score.

50% percentile inside its fair comparison set

56.2 ndcgRaw benchmark value

Multilingual14 benchmarks54%

Text Arena · Chinese

AR · Multilingual · Human

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

Rank #122

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,394
Percentile: 59%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: chinese. Source rank: #148. Votes: 1254. Organization: alibaba. License: Apache 2.0.

59% percentile inside its fair comparison set

1,394Raw benchmark valueCI 1,377 - 1,410

Text Arena · French

AR · Multilingual · Human

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

Rank #124

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,374
Percentile: 43.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: french. Source rank: #147. Votes: 326. Organization: alibaba. License: Apache 2.0.

43.1% percentile inside its fair comparison set

1,374Raw benchmark valueCI 1,342 - 1,407

Text Arena · German

AR · Multilingual · Human

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

Rank #91

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,382
Percentile: 62%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: german. Source rank: #111. Votes: 727. Organization: alibaba. License: Apache 2.0.

62% percentile inside its fair comparison set

1,382Raw benchmark valueCI 1,360 - 1,404

Text Arena · Japanese

AR · Multilingual · Human

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

Rank #100

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,309
Percentile: 51.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: japanese. Source rank: #123. Votes: 646. Organization: alibaba. License: Apache 2.0.

51.2% percentile inside its fair comparison set

1,309Raw benchmark valueCI 1,285 - 1,333

Text Arena · Korean

AR · Multilingual · Human

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

Rank #110

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,318
Percentile: 47.6%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: korean. Source rank: #134. Votes: 471. Organization: alibaba. License: Apache 2.0.

47.6% percentile inside its fair comparison set

1,318Raw benchmark valueCI 1,290 - 1,346

Text Arena · Russian

AR · Multilingual · Human

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

Rank #129

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,354
Percentile: 55.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: russian. Source rank: #156. Votes: 1857. Organization: alibaba. License: Apache 2.0.

55.7% percentile inside its fair comparison set

1,354Raw benchmark valueCI 1,341 - 1,367

Text Arena · Spanish

AR · Multilingual · Human

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

Rank #112

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,369
Percentile: 48.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: spanish. Source rank: #137. Votes: 421. Organization: alibaba. License: Apache 2.0.

48.1% percentile inside its fair comparison set

1,369Raw benchmark valueCI 1,341 - 1,398

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #119

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,380
Percentile: 60%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: chinese. Source rank: #144. Votes: 1254. Organization: alibaba. License: Apache 2.0.

60% percentile inside its fair comparison set

1,380Raw benchmark valueCI 1,363 - 1,396

Text Arena · French · No Style Control

AR · Multilingual · Human

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

Rank #113

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,368
Percentile: 48.1%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: french. Source rank: #134. Votes: 326. Organization: alibaba. License: Apache 2.0.

48.1% percentile inside its fair comparison set

1,368Raw benchmark valueCI 1,336 - 1,401

Text Arena · German · No Style Control

AR · Multilingual · Human

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

Rank #83

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,386
Percentile: 65.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: german. Source rank: #100. Votes: 727. Organization: alibaba. License: Apache 2.0.

65.4% percentile inside its fair comparison set

1,386Raw benchmark valueCI 1,364 - 1,408

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

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

Rank #90

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,308
Percentile: 56.2%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: japanese. Source rank: #111. Votes: 646. Organization: alibaba. License: Apache 2.0.

56.2% percentile inside its fair comparison set

1,308Raw benchmark valueCI 1,284 - 1,332

Text Arena · Korean · No Style Control

AR · Multilingual · Human

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

Rank #100

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,314
Percentile: 52.4%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: korean. Source rank: #120. Votes: 471. Organization: alibaba. License: Apache 2.0.

52.4% percentile inside its fair comparison set

1,314Raw benchmark valueCI 1,286 - 1,342

Text Arena · Russian · No Style Control

AR · Multilingual · Human

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

Rank #126

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,338
Percentile: 56.7%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: russian. Source rank: #153. Votes: 1857. Organization: alibaba. License: Apache 2.0.

56.7% percentile inside its fair comparison set

1,338Raw benchmark valueCI 1,325 - 1,351

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

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

Rank #108

verified runtimeexact alias

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 50%
Last updated: recent
Eligibility: headline eligible

Parsed from Arena leaderboard dataset row `qwen3-235b-a22b`. Category: spanish. Source rank: #130. Votes: 421. Organization: alibaba. License: Apache 2.0.

50% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,334 - 1,391