Gemma 4 31B (Reasoning)
Output 35 tokens/sFirst token 0.9 sContext 256K tokens
Top 5 cheapest providers (Adjusted $/task)
The same list price can give a different adjusted $/task (caching, token efficiency) — click a price for its inputs.
Within the active global provider, residency and confidentiality filters.
| # | Provider | Platform | Raw input $/1M | Raw output $/1M | Adjusted $/task |
|---|---|---|---|---|---|
| 1 | Chutes | Chutes | $0.120 | $0.370 | |
| 2 | Chutes | OpenRouter | $0.120 | $0.370 | |
| 3 | DeepInfra | OpenRouter | $0.090 | $0.340 | |
| 4 | Parasail | OpenRouter | $0.150 | $0.400 | |
| 5 | Venice | OpenRouter | $0.120 | $0.360 |
Composite
4 of 7 inputs · 2 from the model family6 radar axes: DesignArena's two boards share one34.5
includes −5.3 for its Benchmaxxing signal (from 39.8; why, switch off in Options)
Benchmaxxing signal +5.3, light Benchmaxxing tag, uncertain: The 80 % interval reaches below zero — treat this tag as uncertain · light report →The 80 % interval reaches below zero — treat this tag as uncertain.
AA Coding 43.4Coding Agent v1.4 —AA Intelligence 15.4AA Agentic 6.7Epoch ECI 142.7Software ECI 144.1DesignArena —/—
Radar: percentile among all models measured on each input; a gap means not measured.
Benchmark sheet
32 of 140 registered benchmark versions · bars show the percentile among all models measured on each benchmark.
- Epoch ECI · attached
- Software ECI · attached
Agentic
AA-Briefcase17377 Elo
AA-Briefcase published 2026-09-10 ↗ · Published board — Tests multi-week professional knowledge-work projects with linked tasks and large source collections.
GDPval-AA v236755 Elo
GDPval-AA v2 v2 ↗ · Published board — Tests professional knowledge-work deliverables across occupations using AA's Stirrup harness.
Harvey LAB-AA547.2%
Harvey LAB-AA published 2026-09-10 ↗ · Published board — Tests legal-work deliverables across practice areas on Harvey's private task set. Artificial Analysis' run, graded by one LLM judge against task rubrics — not the same run or scale as Vals AI's HLAB row.
ITBench-AA5937.3%
ITBench-AA published 2026-09-10 ↗ · Published board — Tests root-cause diagnosis from offline Kubernetes incident snapshots.
Terminal-Bench Hard (AA)8336.4%
Terminal-Bench Hard (AA) pinned revision 74221fb ↗ · Published board — Tests a pinned 44-task hard subset of terminal-based work using Terminus 2.
Terminal-Bench v2.1 (AA)4443.4%
Terminal-Bench v2.1 (AA) v2.1 ↗ · Published board — Tests terminal-based work on the 89-task verified refresh using Terminus 2.
Terminal-Bench v4.0 (AA)180.0%
Terminal-Bench v4.0 (AA) v4.0 ↗ · Published board — Tests terminal-based work on the 66-task release using mini-SWE-agent v2.4.6.
τ²-Bench Airline (OpenRouter run)8376.1%
τ²-Bench Airline (OpenRouter run) published 2026-09-15 ↗ · Published board — Multi-turn service agents making tool calls under strict policy constraints. (OpenRouter's own reproducible run, as published on openrouter.ai/benchmarks.)
AA Agentic Index366.70
AA Agentic Index published 2026-09-18 ↗ · Published board — Artificial Analysis publishes one Agentic Index per model, measured on its primary configuration and relayed by OpenRouter's Benchmarks API; the value is attached at family scope on the deterministic representative. Artificial Analysis publishes this board without a version number; we keep the date each result was retained.
Coding
SciCode (AA subproblems) v1.0.14545.5%
SciCode (AA subproblems) v1.0.1 ↗ · Published board — Tests scientific Python programming with scientist-annotated background information.
AA Coding Index4843.4
AA Coding Index published 2026-09-19 ↗ · Published board — Artificial Analysis publishes this board without a version number; we keep the date each result was retained.
Efficiency
GPQA Diamond (OpenRouter run) — measured cost per task84$0.0038
GPQA Diamond (OpenRouter run) — measured cost per task published 2026-09-15 ↗ · Published board — Mean USD OpenRouter actually spent per task while running GPQA Diamond for this configuration.
τ²-Bench Airline (OpenRouter run) — measured cost per task84$0.016
τ²-Bench Airline (OpenRouter run) — measured cost per task published 2026-09-15 ↗ · Published board — Mean USD OpenRouter actually spent per task while running τ²-Bench Airline for this configuration.
Instruction-following
IFBench (AA single-turn)9375.6%
IFBench (AA single-turn) published 2026-09-10 ↗ · Published board — Tests precise single-turn instructions with deterministic rule checks.
Knowledge
Humanity's Last Exam (AA text-only)7523.6%
Humanity's Last Exam (AA text-only) published 2026-09-10 ↗ · Published board — Tests expert-level knowledge on AA's text-only Humanity's Last Exam subset.
AA-Omniscience Index37-47.9
AA-Omniscience Index published 2026-09-10 ↗ · Published board — Tests factual reliability while rewarding correct answers and penalizing hallucinations.
UGI NatInt7834.4
UGI NatInt published 2026-09-10 ↗ · Published board — A private-question evaluation of general knowledge and reasoning across textbook, popular-culture and world-model topics.
Long-context
GDP.pdf (AA)356.0%
GDP.pdf (AA) published 2026-09-10 ↗ · Published board — Tests professional reasoning over long PDFs with AA document preparation and grading.
AA-LCR v1.16669.7%
AA-LCR v1.1 v1.1 ↗ · Published board — Tests reasoning across multiple long documents with corrected answer keys and grading.
Reasoning
AA Intelligence Index6215.4
AA Intelligence Index published 2026-09-19 ↗ · Published board — Artificial Analysis publishes this board without a version number; we keep the date each result was retained.
Science
CritPt (AA)681.4%
CritPt (AA) published 2026-09-10 ↗ · Published board — Tests research-level physics reasoning with Python, symbolic and numerical answers.
GPQA Diamond (AA)7985.7%
GPQA Diamond (AA) published 2026-09-10 ↗ · Published board — Tests graduate-level biology, physics and chemistry knowledge on the Diamond subset.
GPQA Diamond (OpenRouter run)6383.4%
GPQA Diamond (OpenRouter run) published 2026-09-15 ↗ · Published board — Graduate-level science questions that resist retrieval and reward careful reasoning. (OpenRouter's own reproducible run, as published on openrouter.ai/benchmarks.)
Tool-use
AutomationBench-AA v1.0.6334.9%
AutomationBench-AA v1.0.6 ↗ · Published board — Tests multi-app SaaS workflows through REST tools on a held-out AutomationBench split.
EnterpriseOps-Gym-AA1128.3%
EnterpriseOps-Gym-AA published 2026-09-10 ↗ · Published board — Tests enterprise workflows through MCP tools against resettable application environments.
τ²-Bench Telecom (AA)5759.9%
τ²-Bench Telecom (AA) published 2026-09-10 ↗ · Published board — Tests dual-control telecom agents that coordinate tool use with a simulated user.
τ³-Banking (AA) v1.0.14214.8%
τ³-Banking (AA) v1.0.1 ↗ · Published board — Tests banking support agents that retrieve policies and change account state through tools.
Uncensored
UGI Willingness272.50
UGI Willingness published 2026-09-10 ↗ · Published board — A private-question evaluation of how far a model follows challenging instructions before refusing or deviating.
UGI2721.5
UGI published 2026-09-10 ↗ · Published board — A private-question suite combining sensitive-topic knowledge with willingness to follow controversial instructions.
Vision
MMMU Pro (AA)6373.4%
MMMU Pro (AA) published 2026-09-10 ↗ · Published board — Tests multimodal understanding using challenging ten-option questions.
Writing
EQ-Bench Creative Writing v3571,366 Elo
EQ-Bench Creative Writing v3 v3 ↗ · Published board — A LLM-judged creative writing benchmark with vocabulary and GPT-slop controls.
Judgemark v410072.3
Judgemark v4 v4 ↗ · Published board — Meta-evaluation that rates LLM judges on how discriminatively they score fixed creative-writing samples.
EQ-Bench Longform Creative Writing v1.115656.5
EQ-Bench Longform Creative Writing v1.11 ↗ · Published board — LLM-judged benchmark in which models plan and write a short story/novella over 8x 1000-word turns, scored 0-100 across 14 rubric dimensions.
UGI Writing7738.6
UGI Writing published 2026-09-10 ↗ · Published board — A maintainer-owned writing evaluation considering intelligence, style, repetition and length adherence, informed by human preferences.
Unusual results
2 threshold-crossing signals flagged · 32 eligible benchmark families
How flags are calculated
Heuristic screen, not statistical significance: benchmark families are correlated and source uncertainty is unknown. Peer evidence needs ≥ 20 independently measured matched configurations from ≥ 10 distinct model families. A flag needs a directed population z-score of magnitude ≥ 1.5 and a gap of ≥ 1.5 from the leave-one-benchmark-family-out mean z in the same direction, over ≥ 5 other benchmark families.
- unusually weak Harvey LAB-AA published 2026-09-10
Why
Observed: 0.47228 fractionPeer mean 0.79402 · peer sd 0.17289 · n 43 · families 38Directed z -1.861 · baseline z 0.074 · gap -1.935 · profile n 310.47228 fractionmeasuredobserved 2026-09-10artificialanalysis.ai ↗Evidence
Axis: Harvey LAB-AA · published 2026-09-10 · Published boardExact value:0.47228253003329fractionObserved: 2026-09-10T21:47:16.627Z · publication date: not recordedObservation id:aa:cd26a386-4873-46ff-b853-d239050025a2:harveyLab - unusually strong IFBench (AA single-turn) published 2026-09-10
Why
Observed: 0.75578 fractionPeer mean 0.48006 · peer sd 0.17139 · n 450 · families 331Directed z 1.609 · baseline z -0.038 · gap 1.646 · profile n 310.75578 fractionmeasuredobserved 2026-09-10artificialanalysis.ai ↗Evidence
Axis: IFBench (AA single-turn) · published 2026-09-10 · Published boardExact value:0.75578231292517fractionObserved: 2026-09-10T21:47:16.627Z · publication date: not recordedObservation id:aa:cd26a386-4873-46ff-b853-d239050025a2:ifbench
Missing coverage · 111 benchmark versions
No result does not mean a zero, or that the model was never tested. Collection failures and disputed versions retain their distinct status.
Variants / reasoning settings
Artificial Analysis snapshot 2026-09-19 · Data: Artificial Analysis · Data: Epoch AI (CC BY)
| Variant | AA Coding | AA Intelligence |
|---|---|---|
| Gemma 4 31B (Reasoning) | 43.4 | 15.4 |
| Gemma 4 31B (Non-reasoning) | 33.2 | 13.9 |
Token offers by platform · 17 offers (Adjusted $/task)
Click any underlined price to see how it is estimated and where each input comes from. How we calculate adjusted cost.
17 offers within the active global filters; “—” means the catalog is active but no public token price is available.
Chutes (1)
| Chutes | globalTEE | $0.120 raw in $/1M | $0.370 raw out $/1M |
OpenRouter (13)
| Chutes | globalchutes/fp4 | $0.120 raw in $/1M | $0.370 raw out $/1M | |
| DeepInfra | globaldeepinfra/turbo | $0.090 raw in $/1M | $0.340 raw out $/1M | |
| Parasail | globalparasail/fp8 | $0.150 raw in $/1M | $0.400 raw out $/1M | |
| Venice | globalvenice/fp4 | $0.120 raw in $/1M | $0.360 raw out $/1M | |
| CoreWeave | globalcoreweave/fp4 | $0.100 raw in $/1M | $0.340 raw out $/1M | |
| DeepInfra | globaldeepinfra/fp8 | $0.130 raw in $/1M | $0.380 raw out $/1M | |
| Novita | globalnovita/bf16 | $0.140 raw in $/1M | $0.400 raw out $/1M | |
| Together | globaltogether | $0.390 raw in $/1M | $0.970 raw out $/1M | |
| ModelRun | globalmodelrun/fp4 | $0.750 raw in $/1M | $1.00 raw out $/1M | |
| Crusoe | globalcrusoe/bf16 | $0.140 raw in $/1M | $0.400 raw out $/1M | |
| DeepInfra | globaldeepinfra/ultra | $0.270 raw in $/1M | $0.760 raw out $/1M | |
| SambaNova | globalsambanova | $0.380 raw in $/1M | $1.15 raw out $/1M | |
| SiliconFlow | globalsiliconflow/fp8 | $0.750 raw in $/1M | $1.00 raw out $/1M |
TrustedTokens (1)
| TrustedTokens | euEU | $0.340 raw in $/1M | $0.920 raw out $/1M |
AWS Bedrock (1)
| AWS Bedrock | eu-central-1EU | $0.168 raw in $/1M | $0.480 raw out $/1M |
STACKIT (1)
| STACKIT | eu01EU | $0.516 raw in $/1M | $0.745 raw out $/1M |