Model profile · Google

gemma-3-27b-it

Open weightsmid · registry tag 2026 benchmark-derived

Thin verified coverage

Reads as thin verified coverage across the resolved source data.

Visible coverage: 1.4%
Verified coverage: 1.4%
Spread: n/a
Last verified: Jun 20, 2026

textcodevision1 aliases38 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 benchmarks58.7%

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 #130

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,366
Percentile: 60.3%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: overall. Source rank: #157. Votes: 47529. Organization: google. License: Gemma.

60.3% percentile inside its fair comparison set

1,366Raw benchmark valueCI 1,362 - 1,369

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 #114

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,348
Percentile: 65%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: creative_writing. Source rank: #139. Votes: 6394. Organization: google. License: Gemma.

65% percentile inside its fair comparison set

1,348Raw benchmark valueCI 1,340 - 1,355

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 #139

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,374
Percentile: 57.5%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: english. Source rank: #166. Votes: 24829. Organization: google. License: Gemma.

57.5% percentile inside its fair comparison set

1,374Raw benchmark valueCI 1,369 - 1,378

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 #129

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,332
Percentile: 60.6%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: exclude_ties. Source rank: #156. Votes: 33351. Organization: google. License: Gemma.

60.6% percentile inside its fair comparison set

1,332Raw benchmark valueCI 1,327 - 1,337

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 #145

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,365
Percentile: 55.7%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: hard_prompts. Source rank: #174. Votes: 17866. Organization: google. License: Gemma.

55.7% percentile inside its fair comparison set

1,365Raw benchmark valueCI 1,360 - 1,371

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 #152

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,370
Percentile: 53.4%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: hard_prompts_english. Source rank: #181. Votes: 9911. Organization: google. License: Gemma.

53.4% percentile inside its fair comparison set

1,370Raw benchmark valueCI 1,364 - 1,377

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 #140

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,344
Percentile: 57.2%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: instruction_following. Source rank: #168. Votes: 12525. Organization: google. License: Gemma.

57.2% percentile inside its fair comparison set

1,344Raw benchmark valueCI 1,338 - 1,349

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 #136

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,364
Percentile: 55.6%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: longer_query. Source rank: #164. Votes: 6645. Organization: google. License: Gemma.

55.6% percentile inside its fair comparison set

1,364Raw benchmark valueCI 1,356 - 1,371

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 #133

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,358
Percentile: 59.1%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: multi_turn. Source rank: #160. Votes: 7023. Organization: google. License: Gemma.

59.1% percentile inside its fair comparison set

1,358Raw benchmark valueCI 1,351 - 1,366

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

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,358
Percentile: 61.2%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: overall. Source rank: #153. Votes: 47529. Organization: google. License: Gemma.

61.2% percentile inside its fair comparison set

1,358Raw benchmark valueCI 1,355 - 1,362

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 #108

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,345
Percentile: 66.9%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: creative_writing. Source rank: #132. Votes: 6394. Organization: google. License: Gemma.

66.9% percentile inside its fair comparison set

1,345Raw benchmark valueCI 1,337 - 1,352

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 #130

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,374
Percentile: 60.3%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: english. Source rank: #154. Votes: 24829. Organization: google. License: Gemma.

60.3% percentile inside its fair comparison set

1,374Raw benchmark valueCI 1,370 - 1,379

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

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,322
Percentile: 61.2%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: exclude_ties. Source rank: #153. Votes: 33351. Organization: google. License: Gemma.

61.2% percentile inside its fair comparison set

1,322Raw benchmark valueCI 1,317 - 1,327

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 #144

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,340
Percentile: 56%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: hard_prompts. Source rank: #171. Votes: 17866. Organization: google. License: Gemma.

56% percentile inside its fair comparison set

1,340Raw benchmark valueCI 1,335 - 1,345

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 #148

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,350
Percentile: 54.6%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: hard_prompts_english. Source rank: #175. Votes: 9911. Organization: google. License: Gemma.

54.6% percentile inside its fair comparison set

1,350Raw benchmark valueCI 1,343 - 1,356

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 #137

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,322
Percentile: 58.2%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: instruction_following. Source rank: #165. Votes: 12525. Organization: google. License: Gemma.

58.2% percentile inside its fair comparison set

1,322Raw benchmark valueCI 1,316 - 1,327

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 #137

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,333
Percentile: 55.3%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: longer_query. Source rank: #165. Votes: 6645. Organization: google. License: Gemma.

55.3% percentile inside its fair comparison set

1,333Raw benchmark valueCI 1,326 - 1,341

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 #135

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,345
Percentile: 58.5%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: multi_turn. Source rank: #162. Votes: 7023. Organization: google. License: Gemma.

58.5% percentile inside its fair comparison set

1,345Raw benchmark valueCI 1,338 - 1,353

Coding2 benchmarks47.3%

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 #179

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,358
Percentile: 44.4%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: coding. Source rank: #212. Votes: 8073. Organization: google. License: Gemma.

44.4% percentile inside its fair comparison set

1,358Raw benchmark valueCI 1,351 - 1,365

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 #160

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,323
Percentile: 50.3%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: coding. Source rank: #190. Votes: 8073. Organization: google. License: Gemma.

50.3% percentile inside its fair comparison set

1,323Raw benchmark valueCI 1,316 - 1,329

Reasoning / math / science2 benchmarks51.4%

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 #153

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,322
Percentile: 51.6%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: math. Source rank: #184. Votes: 3581. Organization: google. License: Gemma.

51.6% percentile inside its fair comparison set

1,322Raw benchmark valueCI 1,313 - 1,332

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 #154

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,311
Percentile: 51.3%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: math. Source rank: #184. Votes: 3581. Organization: google. License: Gemma.

51.3% percentile inside its fair comparison set

1,311Raw benchmark valueCI 1,302 - 1,321

Professional reasoning18 benchmarks58.6%

Text Arena · Expert

AR · Professional reasoning · Human

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

Rank #153

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,335
Percentile: 44.7%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: expert. Source rank: #184. Votes: 2228. Organization: google. License: Gemma.

44.7% percentile inside its fair comparison set

1,335Raw benchmark valueCI 1,322 - 1,348

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 #119

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,377
Percentile: 62.9%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_business_and_management_and_financial_operations. Source rank: #144. Votes: 6580. Organization: google. License: Gemma.

62.9% percentile inside its fair comparison set

1,377Raw benchmark valueCI 1,369 - 1,384

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 #128

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,328
Percentile: 60.7%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_entertainment_and_sports_and_media. Source rank: #156. Votes: 8098. Organization: google. License: Gemma.

60.7% percentile inside its fair comparison set

1,328Raw benchmark valueCI 1,321 - 1,335

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

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

Rank #124

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,382
Percentile: 58.7%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_legal_and_government. Source rank: #150. Votes: 2718. Organization: google. License: Gemma.

58.7% percentile inside its fair comparison set

1,382Raw benchmark valueCI 1,370 - 1,393

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 #122

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,390
Percentile: 62.5%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_life_and_physical_and_social_science. Source rank: #147. Votes: 7943. Organization: google. License: Gemma.

62.5% percentile inside its fair comparison set

1,390Raw benchmark valueCI 1,383 - 1,397

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

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

Rank #147

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,340
Percentile: 52.6%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_mathematical. Source rank: #177. Votes: 3194. Organization: google. License: Gemma.

52.6% percentile inside its fair comparison set

1,340Raw benchmark valueCI 1,330 - 1,350

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

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

Rank #115

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,402
Percentile: 61.4%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_medicine_and_healthcare. Source rank: #138. Votes: 2361. Organization: google. License: Gemma.

61.4% percentile inside its fair comparison set

1,402Raw benchmark valueCI 1,389 - 1,414

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 #153

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,378
Percentile: 53.2%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_software_and_it_services. Source rank: #182. Votes: 14433. Organization: google. License: Gemma.

53.2% percentile inside its fair comparison set

1,378Raw benchmark valueCI 1,372 - 1,384

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 #122

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,355
Percentile: 62.7%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_writing_and_literature_and_language. Source rank: #148. Votes: 10691. Organization: google. License: Gemma.

62.7% percentile inside its fair comparison set

1,355Raw benchmark valueCI 1,348 - 1,361

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

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

Rank #151

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,303
Percentile: 45.5%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: expert. Source rank: #180. Votes: 2228. Organization: google. License: Gemma.

45.5% percentile inside its fair comparison set

1,303Raw benchmark valueCI 1,290 - 1,316

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 #114

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,360
Percentile: 64.5%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_business_and_management_and_financial_operations. Source rank: #136. Votes: 6580. Organization: google. License: Gemma.

64.5% percentile inside its fair comparison set

1,360Raw benchmark valueCI 1,352 - 1,367

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 #126

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,320
Percentile: 61.3%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_entertainment_and_sports_and_media. Source rank: #152. Votes: 8098. Organization: google. License: Gemma.

61.3% percentile inside its fair comparison set

1,320Raw benchmark valueCI 1,314 - 1,327

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 #113

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,375
Percentile: 62.4%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_legal_and_government. Source rank: #136. Votes: 2718. Organization: google. License: Gemma.

62.4% percentile inside its fair comparison set

1,375Raw benchmark valueCI 1,363 - 1,386

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 #112

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,383
Percentile: 65.6%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_life_and_physical_and_social_science. Source rank: #133. Votes: 7943. Organization: google. License: Gemma.

65.6% percentile inside its fair comparison set

1,383Raw benchmark valueCI 1,376 - 1,390

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

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

Rank #149

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,327
Percentile: 51.9%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_mathematical. Source rank: #177. Votes: 3194. Organization: google. License: Gemma.

51.9% percentile inside its fair comparison set

1,327Raw benchmark valueCI 1,317 - 1,337

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 #106

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,390
Percentile: 64.4%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_medicine_and_healthcare. Source rank: #128. Votes: 2361. Organization: google. License: Gemma.

64.4% percentile inside its fair comparison set

1,390Raw benchmark valueCI 1,377 - 1,402

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 #147

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,356
Percentile: 55.1%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_software_and_it_services. Source rank: #174. Votes: 14433. Organization: google. License: Gemma.

55.1% percentile inside its fair comparison set

1,356Raw benchmark valueCI 1,351 - 1,361

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 #115

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,347
Percentile: 64.8%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: industry_writing_and_literature_and_language. Source rank: #140. Votes: 10691. Organization: google. License: Gemma.

64.8% percentile inside its fair comparison set

1,347Raw benchmark valueCI 1,341 - 1,353

Vision understanding20 benchmarks22.9%

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 #61

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,159
Percentile: 45%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: overall. Source rank: #78. Votes: 17406. Organization: google. License: Gemma.

45% percentile inside its fair comparison set

1,159Raw benchmark valueCI 1,151 - 1,168

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 #25

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,110
Percentile: 7.7%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: captioning. Source rank: #27. Votes: 273. Organization: google. License: Gemma.

7.7% percentile inside its fair comparison set

1,110Raw benchmark valueCI 1,073 - 1,147

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 #48

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,176
Percentile: 14.5%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: creative_writing_vision. Source rank: #60. Votes: 630. Organization: google. License: Gemma.

14.5% percentile inside its fair comparison set

1,176Raw benchmark valueCI 1,151 - 1,201

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 #61

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,165
Percentile: 14.3%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: diagram. Source rank: #77. Votes: 1758. Organization: google. License: Gemma.

14.3% percentile inside its fair comparison set

1,165Raw benchmark valueCI 1,150 - 1,180

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 #62

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,170
Percentile: 44%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: english. Source rank: #77. Votes: 8371. Organization: google. License: Gemma.

44% percentile inside its fair comparison set

1,170Raw benchmark valueCI 1,160 - 1,181

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 #27

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,174
Percentile: 18.8%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: entity_recognition. Source rank: #30. Votes: 326. Organization: google. License: Gemma.

18.8% percentile inside its fair comparison set

1,174Raw benchmark valueCI 1,142 - 1,207

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 #57

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,169
Percentile: 17.6%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: homework. Source rank: #73. Votes: 1067. Organization: google. License: Gemma.

17.6% percentile inside its fair comparison set

1,169Raw benchmark valueCI 1,151 - 1,188

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 #42

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,149
Percentile: 16.3%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: humor. Source rank: #54. Votes: 780. Organization: google. License: Gemma.

16.3% percentile inside its fair comparison set

1,149Raw benchmark valueCI 1,126 - 1,173

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 #59

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,162
Percentile: 17.1%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: ocr. Source rank: #75. Votes: 6337. Organization: google. License: Gemma.

17.1% percentile inside its fair comparison set

1,162Raw benchmark valueCI 1,153 - 1,172

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 #62

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,166
Percentile: 44%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: overall. Source rank: #77. Votes: 17406. Organization: google. License: Gemma.

44% percentile inside its fair comparison set

1,166Raw benchmark valueCI 1,158 - 1,174

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 #24

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,144
Percentile: 11.5%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: captioning. Source rank: #26. Votes: 273. Organization: google. License: Gemma.

11.5% percentile inside its fair comparison set

1,144Raw benchmark valueCI 1,107 - 1,181

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 #47

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,186
Percentile: 16.4%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: creative_writing_vision. Source rank: #60. Votes: 630. Organization: google. License: Gemma.

16.4% percentile inside its fair comparison set

1,186Raw benchmark valueCI 1,162 - 1,211

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 #58

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,158
Percentile: 18.6%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: diagram. Source rank: #73. Votes: 1758. Organization: google. License: Gemma.

18.6% percentile inside its fair comparison set

1,158Raw benchmark valueCI 1,143 - 1,172

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 #57

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,186
Percentile: 48.6%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: english. Source rank: #70. Votes: 8371. Organization: google. License: Gemma.

48.6% percentile inside its fair comparison set

1,186Raw benchmark valueCI 1,175 - 1,196

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 #26

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,199
Percentile: 21.9%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: entity_recognition. Source rank: #28. Votes: 326. Organization: google. License: Gemma.

21.9% percentile inside its fair comparison set

1,199Raw benchmark valueCI 1,167 - 1,232

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 #58

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,164
Percentile: 16.2%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: homework. Source rank: #73. Votes: 1067. Organization: google. License: Gemma.

16.2% percentile inside its fair comparison set

1,164Raw benchmark valueCI 1,146 - 1,183

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 #43

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,164
Percentile: 14.3%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: humor. Source rank: #56. Votes: 780. Organization: google. License: Gemma.

14.3% percentile inside its fair comparison set

1,164Raw benchmark valueCI 1,141 - 1,187

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 #60

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,159
Percentile: 15.7%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: ocr. Source rank: #75. Votes: 6337. Organization: google. License: Gemma.

15.7% percentile inside its fair comparison set

1,159Raw benchmark valueCI 1,149 - 1,169

Vision Arena · Creative Writing

AR · Vision understanding · Human

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

Rank #24

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,175
Percentile: 28.1%
Last updated: archived
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: creative_writing. Source rank: #25. Votes: 732. Organization: google. License: Gemma.

28.1% percentile inside its fair comparison set

1,175Raw benchmark valueCI 1,153 - 1,197

Vision Arena · Creative Writing · 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 #24

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,182
Percentile: 28.1%
Last updated: archived
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: creative_writing. Source rank: #25. Votes: 732. Organization: google. License: Gemma.

28.1% percentile inside its fair comparison set

1,182Raw benchmark valueCI 1,161 - 1,204

Multilingual16 benchmarks49.6%

Text Arena · Chinese

AR · Multilingual · Human

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

Rank #137

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,373
Percentile: 53.9%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: chinese. Source rank: #166. Votes: 2290. Organization: google. License: Gemma.

53.9% percentile inside its fair comparison set

1,373Raw benchmark valueCI 1,361 - 1,386

Text Arena · French

AR · Multilingual · Human

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

Rank #116

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,389
Percentile: 46.8%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: french. Source rank: #139. Votes: 600. Organization: google. License: Gemma.

46.8% percentile inside its fair comparison set

1,389Raw benchmark valueCI 1,364 - 1,414

Text Arena · German

AR · Multilingual · Human

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

Rank #104

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,370
Percentile: 56.5%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: german. Source rank: #126. Votes: 1163. Organization: google. License: Gemma.

56.5% percentile inside its fair comparison set

1,370Raw benchmark valueCI 1,352 - 1,388

Text Arena · Japanese

AR · Multilingual · Human

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

Rank #101

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,309
Percentile: 50.7%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: japanese. Source rank: #124. Votes: 951. Organization: google. License: Gemma.

50.7% percentile inside its fair comparison set

1,309Raw benchmark valueCI 1,289 - 1,329

Text Arena · Korean

AR · Multilingual · Human

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

Rank #101

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,326
Percentile: 51.9%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: korean. Source rank: #124. Votes: 866. Organization: google. License: Gemma.

51.9% percentile inside its fair comparison set

1,326Raw benchmark valueCI 1,305 - 1,347

Text Arena · Russian

AR · Multilingual · Human

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

Rank #117

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,368
Percentile: 59.9%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: russian. Source rank: #143. Votes: 2989. Organization: google. License: Gemma.

59.9% percentile inside its fair comparison set

1,368Raw benchmark valueCI 1,358 - 1,378

Text Arena · Spanish

AR · Multilingual · Human

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

Rank #128

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,350
Percentile: 40.7%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: spanish. Source rank: #156. Votes: 774. Organization: google. License: Gemma.

40.7% percentile inside its fair comparison set

1,350Raw benchmark valueCI 1,328 - 1,371

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #140

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,347
Percentile: 52.9%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: chinese. Source rank: #167. Votes: 2290. Organization: google. License: Gemma.

52.9% percentile inside its fair comparison set

1,347Raw benchmark valueCI 1,334 - 1,359

Text Arena · French · No Style Control

AR · Multilingual · Human

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

Rank #111

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,371
Percentile: 49.1%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: french. Source rank: #132. Votes: 600. Organization: google. License: Gemma.

49.1% percentile inside its fair comparison set

1,371Raw benchmark valueCI 1,346 - 1,396

Text Arena · German · No Style Control

AR · Multilingual · Human

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

Rank #100

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,362
Percentile: 58.2%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: german. Source rank: #122. Votes: 1163. Organization: google. License: Gemma.

58.2% percentile inside its fair comparison set

1,362Raw benchmark valueCI 1,345 - 1,380

Text Arena · Japanese · No Style Control

AR · Multilingual · Human

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

Rank #103

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,287
Percentile: 49.8%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: japanese. Source rank: #125. Votes: 951. Organization: google. License: Gemma.

49.8% percentile inside its fair comparison set

1,287Raw benchmark valueCI 1,267 - 1,306

Text Arena · Korean · No Style Control

AR · Multilingual · Human

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

Rank #105

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,309
Percentile: 50%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: korean. Source rank: #125. Votes: 866. Organization: google. License: Gemma.

50% percentile inside its fair comparison set

1,309Raw benchmark valueCI 1,288 - 1,330

Text Arena · Russian · No Style Control

AR · Multilingual · Human

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

Rank #121

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,349
Percentile: 58.5%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: russian. Source rank: #147. Votes: 2989. Organization: google. License: Gemma.

58.5% percentile inside its fair comparison set

1,349Raw benchmark valueCI 1,338 - 1,359

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

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

Rank #123

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,341
Percentile: 43%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: spanish. Source rank: #150. Votes: 774. Organization: google. License: Gemma.

43% percentile inside its fair comparison set

1,341Raw benchmark valueCI 1,319 - 1,362

Vision Arena · Chinese

AR · Multilingual · Human

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

Rank #51

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,169
Percentile: 35.1%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: chinese. Source rank: #65. Votes: 735. Organization: google. License: Gemma.

35.1% percentile inside its fair comparison set

1,169Raw benchmark valueCI 1,144 - 1,195

Vision Arena · Chinese · No Style Control

AR · Multilingual · Human

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

Rank #50

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,155
Percentile: 36.4%
Last updated: recent
Eligibility: benchmark_derived_model

Parsed from Arena leaderboard dataset row `gemma-3-27b-it`. Category: chinese. Source rank: #64. Votes: 735. Organization: google. License: Gemma.

36.4% percentile inside its fair comparison set

1,155Raw benchmark valueCI 1,130 - 1,179

Source links and registry checks

official

Arena

Jun 20, 2026

gemma-3-27b-it

Current snapshot.

Source-linked scores by benchmark

Source links and registry checks

Loading model evidence.

gemma-3-27b-it

Current snapshot.

Source-linked scores by benchmark

Source links and registry checks