DeepSeek V4.1 Flash (Reasoning, Max Effort)
Output 221 tokens/sFirst token 0.8 sContext 1M 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 | DeepInfra | OpenRouter | $0.140 | $0.420 | |
| 2 | Relace | OpenRouter | $0.130 | $0.520 | |
| 3 | Morph | OpenRouter | $0.135 | $0.540 | |
| 4 | TrustedTokens | TrustedTokens | $0.170 | $0.340 | |
| 5 | Wafer | OpenRouter | $0.200 | $0.600 |
Composite
1 of 7 inputs6 radar axes: DesignArena's two boards share one71.3
includes −6.7 for its Benchmaxxing signal (from 78.0; why, switch off in Options)
dominance-adjusted from 87.4: a better-measured model that is at least as good on each of these inputs ranks above it
Benchmaxxing signal +6.7, medium Benchmaxxing tag, uncertain: Based on only 7 comparisons and the 80 % interval reaches below zero — treat this tag as uncertain · medium report →Based on only 7 comparisons and the 80 % interval reaches below zero — treat this tag as uncertain.
AA Coding —Coding Agent v1.4 —AA Intelligence 39.5AA Agentic —Epoch ECI —Software ECI —DesignArena —/—
Radar: percentile among all models measured on each input; a gap means not measured.
Benchmark sheet
19 of 140 registered benchmark versions · bars show the percentile among all models measured on each benchmark.
Agentic
AA-Briefcase821,424 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 v2951,632 Elo
GDPval-AA v2 v2 ↗ · Published board — Tests professional knowledge-work deliverables across occupations using AA's Stirrup harness.
Terminal-Bench v4.0 (AA)8626.8%
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)8776.7%
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.)
Coding
SciCode (AA subproblems) v1.0.16851.9%
SciCode (AA subproblems) v1.0.1 ↗ · Published board — Tests scientific Python programming with scientist-annotated background information.
Efficiency
GPQA Diamond (OpenRouter run) — measured cost per task46$0.032
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 task68$0.028
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.
Knowledge
Humanity's Last Exam (AA text-only)9039.2%
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 Index76-5.30
AA-Omniscience Index published 2026-09-10 ↗ · Published board — Tests factual reliability while rewarding correct answers and penalizing hallucinations.
Long-context
GDP.pdf (AA)5812.8%
GDP.pdf (AA) published 2026-09-10 ↗ · Published board — Tests professional reasoning over long PDFs with AA document preparation and grading.
AA-LCR v1.19984.0%
AA-LCR v1.1 v1.1 ↗ · Published board — Tests reasoning across multiple long documents with corrected answer keys and grading.
Math
ArXivMath 08/2026 (MathArena) v2026-08043.9%
ArXivMath 08/2026 (MathArena) v2026-08 ↗ · Published board — Research-level math problems with a checkable final answer, drawn from arXiv papers submitted in August 2026, so they postdate most training data.
BrokenArXiv 08/2026 (MathArena) v2026-08054.8%
BrokenArXiv 08/2026 (MathArena) v2026-08 ↗ · Published board — Plausible but false statements taken from August 2026 arXiv papers; a model scores by refusing to prove them and saying the statement is false as written.
Reasoning
AA Intelligence Index9439.5
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.
Safety/Alignment
BullshitBench V1 (clear pushback)7241.8%
BullshitBench V1 (clear pushback) published 2026-09-10 ↗ · Published board — Whether a model rejects the broken premise of 55 deliberately nonsensical prompts (V1 question set) instead of answering them confidently.
BullshitBench V2 (clear pushback)7757.0%
BullshitBench V2 (clear pushback) published 2026-09-10 ↗ · Published board — Whether a model rejects the broken premise of 100 deliberately nonsensical prompts (V2 question set) instead of answering them confidently.
Science
CritPt (AA)8914.3%
CritPt (AA) published 2026-09-10 ↗ · Published board — Tests research-level physics reasoning with Python, symbolic and numerical answers.
GPQA Diamond (OpenRouter run)8990.2%
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 v1.0.610068.9%
AutomationBench-AA v1.0.6 ↗ · Published board — Tests multi-app SaaS workflows through REST tools on a held-out AutomationBench split.
Vision
MMMU Pro (AA)7777.0%
MMMU Pro (AA) published 2026-09-10 ↗ · Published board — Tests multimodal understanding using challenging ten-option questions.
Missing coverage · 125 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 · 19 offers (Adjusted $/task)
Click any underlined price to see how it is estimated and where each input comes from. How we calculate adjusted cost.
19 offers within the active global filters; “—” means the catalog is active but no public token price is available.
OpenRouter (16)
| DeepInfra | globaldeepinfra/fp8 | $0.140 raw in $/1M | $0.420 raw out $/1M | |
| Morph | globalmorph/fp8 | $0.135 raw in $/1M | $0.540 raw out $/1M | |
| Wafer | globalwafer | $0.200 raw in $/1M | $0.600 raw out $/1M | |
| Fireworks | globalfireworks | $0.220 raw in $/1M | $0.660 raw out $/1M | |
| Sail Research | globalsail-research/fp4 | $0.200 raw in $/1M | $0.600 raw out $/1M | |
| Novita | globalnovita/fp8 | $0.285 raw in $/1M | $1.14 raw out $/1M | |
| Makora | globalmakora | $0.300 raw in $/1M | $1.20 raw out $/1M | |
| DigitalOcean | globaldigitalocean | $0.300 raw in $/1M | $1.20 raw out $/1M | |
| Together | globaltogether | $0.300 raw in $/1M | $1.20 raw out $/1M | |
| SiliconFlow | globalsiliconflow/fp8 | $0.300 raw in $/1M | $1.20 raw out $/1M | |
| BaseTen | globalbaseten/fp8 | $0.300 raw in $/1M | $1.20 raw out $/1M | |
| Modal | globalmodal | $0.300 raw in $/1M | $1.20 raw out $/1M | |
| Phala | globalphalaTEE | $0.345 raw in $/1M | $1.38 raw out $/1M | |
| Venice | globalvenice/fp8 | $0.375 raw in $/1M | $1.50 raw out $/1M | |
| Relace | globalrelace/fp4 | $0.130 raw in $/1M | $0.520 raw out $/1M | |
| Parasail | globalparasail/fp8 | $0.300 raw in $/1M | $1.20 raw out $/1M |
TrustedTokens (1)
| TrustedTokens | euEU | $0.170 raw in $/1M | $0.340 raw out $/1M |
TensorX (1)
| TensorX | euEU | $0.500 raw in $/1M | $1.50 raw out $/1M |
Nebius (1)
| Nebius | us | $0.300 raw in $/1M | $1.20 raw out $/1M |