Qwen3 235B A22B 2507 Instruct
Context 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 | TensorX | TensorX | $0.070 | $0.460 | |
| 2 | Parasail | OpenRouter | $0.140 | $0.800 | |
| 3 | DeepInfra | OpenRouter | $0.090 | $0.550 | |
| 4 | Novita | OpenRouter | $0.090 | $0.580 | |
| 5 | Venice | OpenRouter | $0.150 | $0.750 |
Composite
1 of 7 inputs6 radar axes: DesignArena's two boards share one20.8
dominance-adjusted from 26.3: a better-measured model that is at least as good on each of these inputs ranks above it
AA Coding —Coding Agent v1.4 —AA Intelligence 12.0AA Agentic —Epoch ECI —Software ECI —DesignArena —/—
Radar: percentile among all models measured on each input; a gap means not measured.
Benchmark sheet
18 of 140 registered benchmark versions · bars show the percentile among all models measured on each benchmark.
Agentic
Terminal-Bench Hard (AA)5215.2%
Terminal-Bench Hard (AA) pinned revision 74221fb ↗ · Published board — Tests a pinned 44-task hard subset of terminal-based work using Terminus 2.
τ²-Bench Airline (OpenRouter run)1945.9%
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.)
Efficiency
GPQA Diamond (OpenRouter run) — measured cost per task86$0.0036
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 task81$0.018
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)5446.1%
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)5811.1%
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 Index42-44.0
AA-Omniscience Index published 2026-09-10 ↗ · Published board — Tests factual reliability while rewarding correct answers and penalizing hallucinations.
UGI NatInt8436.5
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
AA-LCR v1.13533.9%
AA-LCR v1.1 v1.1 ↗ · Published board — Tests reasoning across multiple long documents with corrected answer keys and grading.
Math
AIME 2025 (AA) v20256671.7%
AIME 2025 (AA) v2025 ↗ · Published board — Advanced mathematical problem solving on AIME I and II 2025.
Reasoning
AA Intelligence Index5212.0
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)240.0%
CritPt (AA) published 2026-09-10 ↗ · Published board — Tests research-level physics reasoning with Python, symbolic and numerical answers.
GPQA Diamond (AA)5875.3%
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)3374.4%
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
τ²-Bench Telecom (AA)4133.3%
τ²-Bench Telecom (AA) published 2026-09-10 ↗ · Published board — Tests dual-control telecom agents that coordinate tool use with a simulated user.
Uncensored
UGI Willingness513.20
UGI Willingness published 2026-09-10 ↗ · Published board — A private-question evaluation of how far a model follows challenging instructions before refusing or deviating.
UGI6835.2
UGI published 2026-09-10 ↗ · Published board — A private-question suite combining sensitive-topic knowledge with willingness to follow controversial instructions.
Writing
EQ-Bench Longform Creative Writing v1.116057.4
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 Writing9149.0
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
1 threshold-crossing signal flagged · 19 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 strong UGI Writing published 2026-09-10
Why
Observed: 48.97 pointsPeer mean 31.68366 · peer sd 10.68331 · n 71 · families 70Directed z 1.618 · baseline z -0.023 · gap 1.641 · profile n 1848.97 pointsmeasuredobserved 2026-09-10huggingface.co ↗Evidence
Axis: UGI Writing · published 2026-09-10 · Published boardExact value:48.97pointsObserved: 2026-09-10T22:04:13Z · publication date: not recordedObservation id:public:bab6c2f531028ab3a900e140
Missing coverage · 126 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.
Token offers by platform · 11 offers (Adjusted $/task)
Click any underlined price to see how it is estimated and where each input comes from. How we calculate adjusted cost.
11 offers within the active global filters; “—” means the catalog is active but no public token price is available.
TensorX (1)
| TensorX | euEU | $0.070 raw in $/1M | $0.460 raw out $/1M |
OpenRouter (6)
| Parasail | globalparasail/fp8 | $0.140 raw in $/1M | $0.800 raw out $/1M | |
| DeepInfra | globaldeepinfra/fp8 | $0.090 raw in $/1M | $0.550 raw out $/1M | |
| Novita | globalnovita/fp8 | $0.090 raw in $/1M | $0.580 raw out $/1M | |
| Venice | globalvenice/fp8 | $0.150 raw in $/1M | $0.750 raw out $/1M | |
| Nebius | globalnebius/fp8 | $0.200 raw in $/1M | $0.600 raw out $/1M | |
| globalgoogle-vertex/us-south1 | $0.220 raw in $/1M | $0.880 raw out $/1M |
Nebius (1)
| Nebius | euEU | $0.200 raw in $/1M | $0.600 raw out $/1M |
Google Vertex AI (1)
| Google Vertex AI | global | $0.220 raw in $/1M | $0.880 raw out $/1M |
AWS Bedrock (1)
| AWS Bedrock | eu-central-1EU | $0.290 raw in $/1M | $1.16 raw out $/1M |
Scaleway (1)
| Scaleway | euEU | $0.860 raw in $/1M | $2.58 raw out $/1M |