Intelligence Index
AA · Chat / text · Combined
It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.
Rank #229 · Source label: Gemma 4 31B (Non-reasoning)
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Artificial Analysis
- Raw value
- 14
- Percentile
- 58.9%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.
58.9% percentile inside its fair comparison set14Raw benchmark value
AA-Omniscience accuracy
AA · Chat / text · Objective
It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.
Rank #292 · Source label: Gemma 4 31B (Non-reasoning)
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Artificial Analysis
- Raw value
- 16.6%
- Percentile
- 35.8%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.
35.8% percentile inside its fair comparison set16.6%Raw benchmark value
AA-Omniscience non-hallucination
AA · Chat / text · Objective
It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.
Rank #265 · Source label: Gemma 4 31B (Reasoning)
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Artificial Analysis
- Raw value
- 15%
- Percentile
- 41.7%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.
41.7% percentile inside its fair comparison set15%Raw benchmark value
IFBench
AA · Chat / text · Objective
It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.
Rank #95 · Source label: Gemma 4 31B (Non-reasoning)
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Artificial Analysis
- Raw value
- 53.5%
- Percentile
- 73%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Artificial Analysis public leaderboard field `ifbench`.
73% percentile inside its fair comparison set53.5%Raw benchmark value
Input price
AA · Chat / text · Speed / cost
It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.
Rank #73 · Source label: Gemma 4 31B (Non-reasoning)
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Artificial Analysis
- Raw value
- $0.1 /1M input tokens
- Percentile
- 79.7%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.
79.7% percentile inside its fair comparison set$0.1 /1M input tokensRaw benchmark value
Output price
AA · Chat / text · Speed / cost
It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.
Rank #67 · Source label: Gemma 4 31B (Non-reasoning)
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Artificial Analysis
- Raw value
- $0.4 /1M output tokens
- Percentile
- 81.4%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.
81.4% percentile inside its fair comparison set$0.4 /1M output tokensRaw benchmark value
Output Speed
AA · Chat / text · Speed / cost
It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.
Rank #251 · Source label: Gemma 4 31B (Reasoning)
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Artificial Analysis
- Raw value
- 35.4 tokens/s
- Percentile
- 4.6%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.
4.6% percentile inside its fair comparison set35.4 tokens/sRaw benchmark value
Time to first token
AA · Chat / text · Speed / cost
It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.
Rank #87 · Source label: Gemma 4 31B (Non-reasoning)
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Artificial Analysis
- Raw value
- 2.02s
- Percentile
- 67.4%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.
67.4% percentile inside its fair comparison set2.02sRaw benchmark value
Time to first answer token
AA · Chat / text · Speed / cost
It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.
Rank #217 · Source label: Gemma 4 31B (Reasoning)
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Artificial Analysis
- Raw value
- 50.11s
- Percentile
- 17.6%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.
17.6% percentile inside its fair comparison set50.11sRaw benchmark value
Text Arena
AR · Chat / text · Human
It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.
Rank #69
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,453
- Percentile
- 81.9%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: overall. Source rank: #76. Votes: 6133. Organization: google. License: Apache 2.0.
81.9% percentile inside its fair comparison set1,453Raw benchmark valueCI 1,445 - 1,460
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 #78
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,420
- Percentile
- 79.4%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: creative_writing. Source rank: #86. Votes: 971. Organization: google. License: Apache 2.0.
79.4% percentile inside its fair comparison set1,420Raw benchmark valueCI 1,401 - 1,439
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 #69
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,463
- Percentile
- 81.9%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: english. Source rank: #75. Votes: 2325. Organization: google. License: Apache 2.0.
81.9% percentile inside its fair comparison set1,463Raw benchmark valueCI 1,451 - 1,475
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 #66
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,456
- Percentile
- 82.7%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: exclude_ties. Source rank: #73. Votes: 4214. Organization: google. License: Apache 2.0.
82.7% percentile inside its fair comparison set1,456Raw benchmark valueCI 1,446 - 1,467
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 #67
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,476
- Percentile
- 82.4%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: hard_prompts. Source rank: #73. Votes: 3574. Organization: google. License: Apache 2.0.
82.4% percentile inside its fair comparison set1,476Raw benchmark valueCI 1,466 - 1,485
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 #61
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,486
- Percentile
- 84%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: hard_prompts_english. Source rank: #64. Votes: 1332. Organization: google. License: Apache 2.0.
84% percentile inside its fair comparison set1,486Raw benchmark valueCI 1,471 - 1,502
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 #55
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,453
- Percentile
- 85.6%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: instruction_following. Source rank: #59. Votes: 1802. Organization: google. License: Apache 2.0.
85.6% percentile inside its fair comparison set1,453Raw benchmark valueCI 1,439 - 1,467
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 #56
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,470
- Percentile
- 84.5%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: longer_query. Source rank: #60. Votes: 1798. Organization: google. License: Apache 2.0.
84.5% percentile inside its fair comparison set1,470Raw benchmark valueCI 1,457 - 1,484
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 #63
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,465
- Percentile
- 83.4%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: multi_turn. Source rank: #70. Votes: 1127. Organization: google. License: Apache 2.0.
83.4% percentile inside its fair comparison set1,465Raw benchmark valueCI 1,448 - 1,482
Text Arena · No Style Control
AR · Chat / text · Human
It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.
Rank #62
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,443
- Percentile
- 83.8%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: overall. Source rank: #67. Votes: 6133. Organization: google. License: Apache 2.0.
83.8% percentile inside its fair comparison set1,443Raw benchmark valueCI 1,436 - 1,451
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 #67
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,416
- Percentile
- 82.4%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: creative_writing. Source rank: #73. Votes: 971. Organization: google. License: Apache 2.0.
82.4% percentile inside its fair comparison set1,416Raw benchmark valueCI 1,397 - 1,435
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 #63
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,451
- Percentile
- 83.5%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: english. Source rank: #66. Votes: 2325. Organization: google. License: Apache 2.0.
83.5% percentile inside its fair comparison set1,451Raw benchmark valueCI 1,439 - 1,463
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 #57
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,442
- Percentile
- 85.1%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: exclude_ties. Source rank: #61. Votes: 4214. Organization: google. License: Apache 2.0.
85.1% percentile inside its fair comparison set1,442Raw benchmark valueCI 1,431 - 1,453
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 #75
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,448
- Percentile
- 80.3%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: hard_prompts. Source rank: #79. Votes: 3574. Organization: google. License: Apache 2.0.
80.3% percentile inside its fair comparison set1,448Raw benchmark valueCI 1,438 - 1,457
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 #67
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,457
- Percentile
- 82.4%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: hard_prompts_english. Source rank: #70. Votes: 1332. Organization: google. License: Apache 2.0.
82.4% percentile inside its fair comparison set1,457Raw benchmark valueCI 1,441 - 1,472
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 #65
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,433
- Percentile
- 83%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: instruction_following. Source rank: #69. Votes: 1802. Organization: google. License: Apache 2.0.
83% percentile inside its fair comparison set1,433Raw benchmark valueCI 1,419 - 1,446
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 #65
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,446
- Percentile
- 82%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: longer_query. Source rank: #68. Votes: 1798. Organization: google. License: Apache 2.0.
82% percentile inside its fair comparison set1,446Raw benchmark valueCI 1,433 - 1,460
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 #61
verified runtimeexact alias
Raw row drilldownsource row, percentile, last updated, eligibility
- Source
- Arena
- Raw value
- 1,452
- Percentile
- 84%
- Last updated
- recent
- Eligibility
- headline eligible
Parsed from Arena leaderboard dataset row `gemma-4-31b`. Category: multi_turn. Source rank: #65. Votes: 1127. Organization: google. License: Apache 2.0.
84% percentile inside its fair comparison set1,452Raw benchmark valueCI 1,435 - 1,469