GPT-6 Astra (max)
Output 56 tokens/sFirst token 199 sContext 1M tokens
Top 2 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 | Azure | OpenRouter | $10.00 | $50.00 | |
| 2 | Azure AI Foundry | Azure AI Foundry | $10.00 | $50.00 |
Composite
7 of 7 inputs · 4 from the model family6 radar axes: DesignArena's two boards share one97.9
AA Coding 76.9Coding Agent v1.4 67.0AA Intelligence 52.8AA Agentic —Epoch ECI 166.3Software ECI 163.6DesignArena 1333/1363
Radar: percentile among all models measured on each input; a gap means not measured.
Benchmark sheet
48 of 140 registered benchmark versions · bars show the percentile among all models measured on each benchmark.
- Epoch ECI · attached
- Software ECI · attached
- DesignArena Web Apps (agentic) · attached
- DesignArena Full-Stack · attached
Agentic
AA-AnalystAgent9051.2%
AA-AnalystAgent published 2026-09-10 ↗ · Published board — Tests analyst tasks using agentic Python execution across fourteen domains.
AA-Briefcase951,562 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 v2911,580 Elo
GDPval-AA v2 v2 ↗ · Published board — Tests professional knowledge-work deliverables across occupations using AA's Stirrup harness.
Terminal-Bench v2.1 (AA)9688.4%
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)9959.1%
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.
ApprenticeBench CUA (NeoCognition)10068.0%
ApprenticeBench CUA (NeoCognition) published 2026-09-14 ↗ · Codex — A computer-use agent operates the Odoo ERP through its screens across 100 sequential accounts-payable tasks with diminishing mentoring.
ApprenticeBench API (NeoCognition)10065.0%
ApprenticeBench API (NeoCognition) published 2026-09-14 ↗ · Codex — An agent completes the same 100 sequential accounts-payable tasks in the Odoo ERP with diminishing mentoring, using dedicated application API calls instead of the screen interface.
Vals Index v2 (Vals AI)9366.6%
Vals Index v2 (Vals AI) v2 ↗ · Published board — GDP-weighted average of agentic model performance across finance, coding and legal tasks.
Finance Agent v2 (Vals Index v2)4053.5%
Finance Agent v2 (Vals Index v2) v2 ↗ · Published board — Multi-step financial reasoning tasks.
Terminal-Bench 2.1 (Vals Index v2)10087.3%
Terminal-Bench 2.1 (Vals Index v2) v2 ↗ · Published board — Command-line interface problem solving, as run by Vals AI.
HLAB — Harvey's Legal Agent Benchmark (Vals Index v2)475.4%
HLAB — Harvey's Legal Agent Benchmark (Vals Index v2) v2 ↗ · Published board — Long-horizon legal work product creation. Vals AI's run of Harvey's Legal Agent Benchmark, reported as accuracy — not comparable with Artificial Analysis' Harvey LAB-AA row.
EBR-bench (Earthborne Rangers, Epoch AI)10076.2%
EBR-bench (Earthborne Rangers, Epoch AI) published 2026-09-18 ↗ · Published board — Learning-capability test: models repeatedly play the obscure campaign board game Earthborne Rangers with note-taking, and the score measures whether results improve across playthroughs.
Coding
Artificial Analysis Coding Agent Index v1.410067.0%
Artificial Analysis Coding Agent Index v1.4 v1.4 ↗ · Codex — The retained AA Coding Agent Index measures coding-agent systems using the earlier three-component implementation.
Artificial Analysis Coding Agent Index v1.510061.6%
Artificial Analysis Coding Agent Index v1.5 v1.5 ↗ · Codex — Measures coding-agent systems on DeepSWE v1.1, Terminal-Bench 4.0 and SWE-Atlas-QnA.
SciCode (AA subproblems) v1.0.18856.5%
SciCode (AA subproblems) v1.0.1 ↗ · Published board — Tests scientific Python programming with scientist-annotated background information.
Vibe Code Bench (Vals Index v2)9489.6%
Vibe Code Bench (Vals Index v2) v2 ↗ · Published board — End-to-end app-building tasks.
Code Migration (Vals Index v2 subset)10068.5%
Code Migration (Vals Index v2 subset) v2 ↗ · Published board — Porting projects to another language, including COBOL modernization.
FrontierCode 1.1 Main (Cognition) v1.1no percentile53.3%
FrontierCode 1.1 Main (Cognition) v1.1 ↗ · codex — Whether a maintainer would merge the agent's pull request, on tasks crafted by open-source maintainers and graded with tests, rubrics and verifiers.
DeepSWE (Datacurve, via Epoch AI)9473.2%
DeepSWE (Datacurve, via Epoch AI) published 2026-09-15 ↗ · mini-swe-agent — Pass rate of coding agents on original, long-horizon software engineering tasks, run by Datacurve with the mini-swe-agent harness.
VulcanBench Frontier v410089.3%
VulcanBench Frontier v4 v4 ↗ · Codex — 23 behavioural-reconstruction tasks: the model must repair a replacement implementation of a legacy program until hidden tests confirm it reproduces the program’s real drift from its written spec; the combined score weights functional correctness 50%, lint/complexity 8.5%, security 8.5% and judged Code quality 33%.
AA Coding Index9676.9
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
ApprenticeBench CUA cost per task (NeoCognition)0$20.51
ApprenticeBench CUA cost per task (NeoCognition) published 2026-09-14 ↗ · Codex — ApprenticeBench's published USD cost per task for each CUA model-harness-effort configuration on the 100-task accounts-payable job.
ApprenticeBench API cost per task (NeoCognition)0$9.62
ApprenticeBench API cost per task (NeoCognition) published 2026-09-14 ↗ · Codex — ApprenticeBench's published USD cost per task for each API-board model-harness-effort configuration on the 100-task accounts-payable job.
Vals Index v2 cost per test (Vals AI)7$19.09
Vals Index v2 cost per test (Vals AI) v2 ↗ · Published board — Vals AI's published USD cost per test for each model on the Vals Index v2.
FrontierCode 1.1 Main cost per rollout (Cognition) v1.1no percentile$4.59
FrontierCode 1.1 Main cost per rollout (Cognition) v1.1 ↗ · codex — Cognition's published mean USD spend per rollout for each FrontierCode 1.1 Main model and reasoning effort.
Knowledge
Humanity's Last Exam (AA text-only)9954.7%
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 Index9943.4
AA-Omniscience Index published 2026-09-10 ↗ · Published board — Tests factual reliability while rewarding correct answers and penalizing hallucinations.
Legal Research Bench (Vals Index v2)5639.4%
Legal Research Bench (Vals Index v2) v2 ↗ · Published board — Case and statute research with citation-backed answers.
SimpleQA Verified (run by Epoch AI)10075.6%
SimpleQA Verified (run by Epoch AI) published 2026-09-16 ↗ · Published board — Short fact-seeking questions answered without tools, testing whether a model knows a fact rather than guessing; Google DeepMind's cleaned version of OpenAI's SimpleQA, run by Epoch AI.
Long-context
GDP.pdf (AA)9931.0%
GDP.pdf (AA) published 2026-09-10 ↗ · Published board — Tests professional reasoning over long PDFs with AA document preparation and grading.
AA-LCR v1.19280.7%
AA-LCR v1.1 v1.1 ↗ · Published board — Tests reasoning across multiple long documents with corrected answer keys and grading.
MLCR-AA8435.0%
MLCR-AA published 2026-09-10 ↗ · Published board — Tests medical-record synthesis and reasoning across long, fragmented case documents.
Math
ArXivMath 06/2026 (MathArena) v2026-0610094.4%
ArXivMath 06/2026 (MathArena) v2026-06 ↗ · Published board — Research-level math problems with a checkable final answer, drawn from arXiv papers submitted in June 2026, so they postdate most training data.
BrokenArXiv 06/2026 (MathArena) v2026-0610099.1%
BrokenArXiv 06/2026 (MathArena) v2026-06 ↗ · Published board — Plausible but false proof statements taken from June 2026 arXiv papers; a model scores by refusing to prove them and saying the statement is false as written.
FrontierMath Tiers 1–3 v2 (Epoch AI)10093.7%
FrontierMath Tiers 1–3 v2 (Epoch AI) v2 ↗ · Published board — Unpublished, expert-written mathematics problems from undergraduate to research level with automatically checkable answers, run by Epoch AI on its private v2 set.
FrontierMath Tier 4 v2 (Epoch AI)9797.6%
FrontierMath Tier 4 v2 (Epoch AI) v2 ↗ · Published board — The hardest, research-level tier of Epoch AI's unpublished FrontierMath problems, run by Epoch AI on its private v2 set.
ArXivMath 08/2026 (MathArena) v2026-0810088.6%
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) v2026-0810081.9%
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
Chess Puzzles (Epoch AI)10072.0%
Chess Puzzles (Epoch AI) published 2026-09-18 ↗ · Published board — Best-move selection on 100 novel chess positions generated programmatically by Epoch AI, each with a single Stockfish-verified best move; probes spatial reasoning and planning.
Mystery Game Puzzles (Epoch AI)10084.0%
Mystery Game Puzzles (Epoch AI) published 2026-09-18 ↗ · Published board — Best-move selection on 100 mid-game positions of a well-known game whose identity Epoch deliberately keeps undisclosed, generated programmatically like Chess Puzzles.
AA Intelligence Index10052.8
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)7447.3%
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)8469.0%
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)10031.7%
CritPt (AA) published 2026-09-10 ↗ · Published board — Tests research-level physics reasoning with Python, symbolic and numerical answers.
GPQA Diamond (AA)10096.1%
GPQA Diamond (AA) published 2026-09-10 ↗ · Published board — Tests graduate-level biology, physics and chemistry knowledge on the Diamond subset.
GPQA Diamond (Epoch AI run)10095.8%
GPQA Diamond (Epoch AI run) published 2026-09-18 ↗ · Published board — Epoch AI's own inspect-ai runs of GPQA Diamond, the 198-question graduate-level science multiple-choice set.
Tool-use
AutomationBench-AA v1.0.69968.5%
AutomationBench-AA v1.0.6 ↗ · Published board — Tests multi-app SaaS workflows through REST tools on a held-out AutomationBench split.
τ³-Banking (AA) v1.0.18841.4%
τ³-Banking (AA) v1.0.1 ↗ · Published board — Tests banking support agents that retrieve policies and change account state through tools.
Excel Modeling Benchmark (Vals Index v2)8471.7%
Excel Modeling Benchmark (Vals Index v2) v2 ↗ · Published board — Building and editing financial models in spreadsheets.
Vision
MMMU Pro (AA)10086.9%
MMMU Pro (AA) published 2026-09-10 ↗ · Published board — Tests multimodal understanding using challenging ten-option questions.
Unusual results
5 threshold-crossing signals flagged · 35 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 Vals Index v2 cost per test (Vals AI)
Why
Observed: 19.08878 USDPeer mean 7.03053 · peer sd 7.84824 · n 31 · families 30Directed z -1.536 · baseline z 1.526 · gap -3.062 · profile n 3419.08878 USDmeasuredobserved 2026-09-18vals.ai ↗Evidence
Axis: Vals Index v2 cost per test (Vals AI) · v2 · Published boardExact value:19.088776USDObserved: 2026-09-18T13:00:43.826192+00:00 · publication date: not recordedObservation id:public:951ce6690d69cd50f95fe408 - unusually strong CritPt (AA) published 2026-09-10
Why
Observed: 0.31714 fractionPeer mean 0.03928 · peer sd 0.07596 · n 522 · families 369Directed z 3.658 · baseline z 1.373 · gap 2.285 · profile n 340.31714 fractionmeasuredobserved 2026-09-10artificialanalysis.ai ↗Evidence
Axis: CritPt (AA) · published 2026-09-10 · Published boardExact value:0.317142857142857fractionObserved: 2026-09-10T21:47:16.627Z · publication date: not recordedObservation id:aa:2f339a97-9a0d-499a-9cb5-e0db665bfa25:critpt - unusually strong AA Intelligence Index published 2026-09-19
Why
Observed: 52.8 pointsPeer mean 15.94317 · peer sd 11.48893 · n 644 · families 483Directed z 3.208 · baseline z 1.386 · gap 1.822 · profile n 3452.8 pointsmeasuredobserved 2026-09-19artificialanalysis.ai ↗Evidence
Axis: AA Intelligence Index · published 2026-09-19 · Published boardExact value:52.8pointsObserved: 2026-09-19 · publication date: not recordedObservation id:legacy:aa_intelligence_index:gpt-6-astra::maxProtocol: inspect · filedata/raw/artificialanalysis.json - unusually strong Terminal-Bench v4.0 (AA)
Why
Observed: 0.59091 fractionPeer mean 0.10006 · peer sd 0.15678 · n 149 · families 109Directed z 3.131 · baseline z 1.415 · gap 1.716 · profile n 340.59091 fractionmeasuredobserved 2026-09-10artificialanalysis.ai ↗Evidence
Axis: Terminal-Bench v4.0 (AA) · v4.0 · Published boardExact value:0.590909090909091fractionObserved: 2026-09-10T21:47:16.627Z · publication date: not recordedObservation id:aa:2f339a97-9a0d-499a-9cb5-e0db665bfa25:terminalbenchV40 - unusually strong Mystery Game Puzzles (Epoch AI) published 2026-09-18
Why
Observed: 0.84 fractionPeer mean 0.31109 · peer sd 0.17549 · n 38 · families 29Directed z 3.014 · baseline z 1.392 · gap 1.622 · profile n 340.84 fractionmeasuredobserved 2026-09-18epoch.ai ↗Evidence
Axis: Mystery Game Puzzles (Epoch AI) · published 2026-09-18 · Published boardExact value:0.84fractionObserved: 2026-09-18T23:49:00Z · publication date: not recordedObservation id:public:55a84368f006dde9d6900b85
Missing coverage · 95 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 |
|---|---|---|
| GPT-6 Astra (high) | 77.1 | 51.0 |
| GPT-6 Astra (max) | 76.9 | 52.8 |
| GPT-6 Astra (medium) | 76.7 | 49.7 |
| GPT-6 Astra (xhigh) | 75.9 | 52.5 |
| GPT-6 Astra (low) | 75.7 | 46.0 |
SUBSCRIPTION PLAN
GitHub Copilot
Current usage-based billing · model token cost is converted to AI Credits at 1 credit = $0.01.
Token offers by platform · 3 offers (Adjusted $/task)
Click any underlined price to see how it is estimated and where each input comes from. How we calculate adjusted cost.
3 offers within the active global filters; “—” means the catalog is active but no public token price is available.
OpenRouter (2)
| Azure | globalazure | $10.00 raw in $/1M | $50.00 raw out $/1M | |
| Azure | globalazure/us | $11.00 raw in $/1M | $55.00 raw out $/1M |
Azure AI Foundry (1)
| Azure AI Foundry | global | $10.00 raw in $/1M | $50.00 raw out $/1M |