# GPT-6 Astra Benchmark Scores & Performance

> GPT-6 Astra by OpenAI scores 83.88/100 overall, ranking #3 out of 954 AI models.

Canonical page: https://benchlm.ai/models/gpt-6-astra

Last updated: October 17, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | OpenAI |
| Source Type | Proprietary |
| Reasoning Type | Reasoning |
| Context Window | 1.05M |
| Overall Score | 83.88/100 |
| Overall Rank | #3 of 954 |

## Family & Coverage

- Family: GPT-6
- Variant: astra (astra)
- Benchmarks covered: 74 of 639
- Sibling models: [GPT-6.1 Sol](/models/gpt-6-1-sol), [GPT-6 Sol](/models/gpt-6-sol), [GPT-6 Luna](/models/gpt-6-luna)
- Related earlier model: [GPT-5.6 Sol](/models/gpt-5-6-sol)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [BrowseComp](/benchmarks/browsecomp) | 91.5% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [OSWorld 2.0](/benchmarks/osworld2) | 72.6% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [Terminal-Bench 4.0](/benchmarks/terminal-bench-4) | 57.90% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [Terminal-Bench-Science 0.1](/benchmarks/terminal-bench-science) | 64.6% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [ExploitGym](/benchmarks/exploitgym) | 42.4% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [Agents' Last Exam](/benchmarks/agentslastexam) | 59.3% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [AutomationBench](/benchmarks/automationbench) | 41.4% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [HLE w/ tools](/benchmarks/hlewithtools) | 57.2% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [GDPval-AA](/benchmarks/gdpvalaanormalized) | 53.6% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/gpt-6-astra) |
| [AA Briefcase](/benchmarks/aabriefcaseelo) | 1570 | [Artificial Analysis: aa-briefcase leaderboard](https://artificialanalysis.ai/evaluations/aa-briefcase) |
| [AA Tau3 Banking](/benchmarks/aatau3banking) | 41.4% | [Artificial Analysis: tau3-banking leaderboard](https://artificialanalysis.ai/evaluations/tau3-banking) |
| [Terminal-Bench 2.1 (Vals)](/benchmarks/terminalbench21) | 87.3% | [Vals AI: Terminal-Bench 2.1 leaderboard](https://www.vals.ai/models/openai_gpt-6-astra) |
| [AA AutomationBench](/benchmarks/aaautomationbench) | 68.5% | [Artificial Analysis: automationbench-aa leaderboard](https://artificialanalysis.ai/evaluations/automationbench-aa) |
| [GDPval-AA](/benchmarks/gdpvalaa) | 1573 | [Artificial Analysis: gdpval-aa leaderboard](https://artificialanalysis.ai/evaluations/gdpval-aa) |
| [AA Agentic Index](/benchmarks/aaagenticindex) | 51.5% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/gpt-6-astra) |
| [GDP.pdf](/benchmarks/aagdppdf) | 31.0% | [Artificial Analysis: gdp-pdf leaderboard](https://artificialanalysis.ai/evaluations/gdp-pdf) |
| [AA-AnalystAgent](/benchmarks/aaanalystagent) | 51.2% | [Artificial Analysis: aa-analyst-agent leaderboard](https://artificialanalysis.ai/evaluations/aa-analyst-agent) |
| [ApprenticeBench](/benchmarks/apprenticebench) | 68% | [NeoCognition: ApprenticeBench GUI results](https://neocognition.io/blog/apprentice-bench/#are-frontier-agents-ready-for-the-job) |
| [AA ITBench](/benchmarks/aaitbench) | 48.6% | [Artificial Analysis: itbench-aa leaderboard](https://artificialanalysis.ai/evaluations/itbench-aa) |
| [aaTerminalBench21](/benchmarks/terminalbench21) | 88.4% | [Artificial Analysis: terminalbench-v2-1 leaderboard](https://artificialanalysis.ai/evaluations/terminalbench-v2-1) |
| [AA Terminal-Bench 4.0](/benchmarks/terminal-bench-4) | 59.1% | [Artificial Analysis: terminalbench-v4-0 leaderboard](https://artificialanalysis.ai/evaluations/terminalbench-v4-0) |
| [CWE-bench v1](/benchmarks/cwe-bench-v1) | 68.0% | [Collinear: CWE-bench v1 leaderboard](https://cwe-bench.com/) |
| [Hermes Index](/benchmarks/hermesindex) | 56.3% | [Nous Research: Hermes Index leaderboard](https://portal.nousresearch.com/bench) |
| [Hermes Bench](/benchmarks/hermesbench) | 73.7% | [Nous Research: Hermes Index leaderboard](https://portal.nousresearch.com/bench) |
| [MazeBench](/benchmarks/mazebench) | 23% | [MazeBench AI leaderboard (ASCII mode)](https://mazebench.com/leaderboard/ai?mode=ascii&tools=1&metric=gems&run=run-2026-09-07T20-06-12-768Z-cba537) |

## Coding Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [DeepSWE](/benchmarks/deepswe) | 74.1% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [FrontierCode 1.1 Main](/benchmarks/frontiercode) | 53.3% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [FrontierCode 1.1 Extended](/benchmarks/frontiercode11extended) | 64.5% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [AA-SciCode](/benchmarks/aascicode) | 56.5% | [Artificial Analysis: scicode leaderboard](https://artificialanalysis.ai/evaluations/scicode) |
| [AA Coding Index](/benchmarks/aacodingindex) | 76.9% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/gpt-6-astra) |
| [FrontierSWE v2](/benchmarks/frontierswev2) | 65.5% | [Proximal: FrontierSWE v2 leaderboard](https://www.frontierswe.com/) |
| [PostTrainBench v1.1](/benchmarks/posttrainbench-v1-1) | 44.3% | [Google: Gemini 4 Argon launch chart](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/) |
| [Bug Hunt Bench](/benchmarks/bug-hunt-bench) | 45.0 fixes | [Bug Hunt Bench public data](https://bughunt.productcompass.pm/data/benchmark.json) |

## Multimodal & Grounded Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [ScreenSpot Pro](/benchmarks/screenspot-pro) | 92.7% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [BenchCAD Vision2Code (tools)](/benchmarks/benchcadvision2codewithtools) | 0.959 | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [AA-MMMU-Pro](/benchmarks/mmmu-pro) | 86.9% | [Artificial Analysis: mmmu-pro leaderboard](https://artificialanalysis.ai/evaluations/mmmu-pro) |

## Reasoning Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [STEMBench](/benchmarks/stembench) | 2.81 | [STEMBench120 October 14, 2026 snapshot audit (Ulam)](https://www.ulam.ai/stembench#full-run-results) |
| [ARC-AGI-1](/benchmarks/arcagi1) | 98.50% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [ARC-AGI-2](/benchmarks/arc-agi-2) | 95% | [ARC Prize: GPT-6 Astra verified results](https://arcprize.org/results/openai-gpt-6-astra) |
| [ARC-AGI-3](/benchmarks/arcagi3) | 62.7% | [ARC Prize: GPT-6 Astra verified results](https://arcprize.org/results/openai-gpt-6-astra) |
| [GeneBench-Pro](/benchmarks/genebenchpro) | 37.8% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [MRCR v2 256K-512K](/benchmarks/mrcr-v2-256k-512k) | 100.0% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [MRCR v2 512K-1M](/benchmarks/mrcr-v2-512k-1m) | 96.3% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [AA-LCR](/benchmarks/lcr) | 80.7% | [Artificial Analysis: artificial-analysis-long-context-reasoning leaderboard](https://artificialanalysis.ai/evaluations/artificial-analysis-long-context-reasoning) |
| [CritPt](/benchmarks/critpt) | 31.7% | [Artificial Analysis: critpt leaderboard](https://artificialanalysis.ai/evaluations/critpt) |
| [MLCR-AA](/benchmarks/aamlcr) | 35.0% | [Artificial Analysis: mlcr-aa leaderboard](https://artificialanalysis.ai/evaluations/mlcr-aa) |
| [GraphWalks BFS 256K–1M](/benchmarks/graphwalks-bfs-256k-1m) | 71.8% | [Google: Gemini 4 Argon launch chart](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/) |

## Knowledge Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [GPQA](/benchmarks/gpqa) | 96% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [GPQA-D](/benchmarks/gpqa-diamond) | 96.0% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [HealthBench (raw)](/benchmarks/healthbench) | 56.9% | [OpenAI GPT-6 Astra system card](https://deploymentsafety.openai.com/gpt-6-astra) |
| [HealthBench (length-adjusted)](/benchmarks/healthbenchlengthadjusted) | 58.3% | [OpenAI GPT-6 Astra system card](https://deploymentsafety.openai.com/gpt-6-astra) |
| [HealthBench Professional](/benchmarks/healthbenchprofessional) | 64.7% | [OpenAI GPT-6 Astra system card](https://deploymentsafety.openai.com/gpt-6-astra) |
| [HealthBench Professional (raw)](/benchmarks/healthbenchprofessionalraw) | 68.2% | [OpenAI GPT-6 Astra system card](https://deploymentsafety.openai.com/gpt-6-astra) |
| [HealthBench Hard](/benchmarks/healthbench-hard) | 36.6% | [OpenAI GPT-6 Astra system card](https://deploymentsafety.openai.com/gpt-6-astra) |
| [Artificial Analysis Intelligence Index](/benchmarks/artificialanalysis) | 52.7% | [Artificial Analysis: artificial-analysis-intelligence-index leaderboard](https://artificialanalysis.ai/evaluations/artificial-analysis-intelligence-index) |
| [AA-GPQA Diamond](/benchmarks/gpqa-diamond) | 96.1% | [Artificial Analysis: gpqa-diamond leaderboard](https://artificialanalysis.ai/evaluations/gpqa-diamond) |
| [AA-HLE](/benchmarks/aahle) | 54.7% | [Artificial Analysis: humanitys-last-exam leaderboard](https://artificialanalysis.ai/evaluations/humanitys-last-exam) |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | 43.4% | [Artificial Analysis: omniscience leaderboard](https://artificialanalysis.ai/evaluations/omniscience) |
| [AA-Omniscience Accuracy](/benchmarks/omniscienceaccuracy) | 62.6% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/gpt-6-astra) |
| [AA-Omniscience Hallucination Rate](/benchmarks/omnisciencehallucinationrate) | 51.3% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/gpt-6-astra) |

## Instruction Following Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [Gray Swan IPI (15 attempts)](/benchmarks/gray-swan-ipi) | 8.5% | [Google: Gemini 4 Argon launch chart](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/) |

## Mathematics Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [FrontierMath v2 (Tier 4)](/benchmarks/frontiermathv2tier4) | 97.600% | [OpenAI: GPT-6 Astra](https://openai.com/index/gpt-6-astra/) |
| [FrontierMath v2 (Tiers 1-3)](/benchmarks/frontiermathv2tiers13) | 93.684% | [Epoch AI FrontierMath v2 leaderboard](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) |

## Other OpenAI Models

- [GPT-6.1 Sol](/models/gpt-6-1-sol) - Score: 84.07
- [GPT-6 Sol](/models/gpt-6-sol) - Score: 78.48
- [GPT-5.6 Sol](/models/gpt-5-6-sol) - Score: 75.32
- [GPT-5.5 Pro](/models/gpt-5-5-pro) - Score: 73.37
- [GPT-5.6 Terra](/models/gpt-5-6-terra) - Score: 70.21
- [GPT-5.5](/models/gpt-5-5) - Score: 69.29
- [GPT-5.4 Pro](/models/gpt-5-4-pro) - Score: 68.6
- [GPT-5.4](/models/gpt-5-4) - Score: 66.7
- [GPT-6 Luna](/models/gpt-6-luna) - Score: 66.66
- [GPT-5.2 Pro](/models/gpt-5-2-pro) - Score: 66.65
- [GPT-5.6 Luna](/models/gpt-5-6-luna) - Score: 64.35
- [GPT-5.3 Codex](/models/gpt-5-3-codex) - Score: 60.48
- [GPT-5.2](/models/gpt-5-2) - Score: 59.86
- [GPT-5.1](/models/gpt-5-1) - Score: 57.21
- [GPT-5.2-Codex](/models/gpt-5-2-codex) - Score: 55.3
- [GPT-5.4 mini](/models/gpt-5-4-mini) - Score: 53.41
- [GPT-5.1-Codex](/models/gpt-5-1-codex) - Score: 50.37
- [GPT-5 mini](/models/gpt-5-mini) - Score: 48.39
- [GPT-5.4 nano](/models/gpt-5-4-nano) - Score: 46.86
- [GPT-5 (medium)](/models/gpt-5-medium) - Score: 45.99
- [o3](/models/o3) - Score: 45.44
- [o3-pro](/models/o3-pro) - Score: 45.38
- [o3-mini](/models/o3-mini) - Score: 39.56
- [o1](/models/o1) - Score: 38.38
- [GPT-4.1](/models/gpt-4-1) - Score: 37.6
- [GPT-5 nano](/models/gpt-5-nano) - Score: 37.37
- [GPT-OSS 120B](/models/gpt-oss-120b) - Score: 37.32
- [o1-preview](/models/o1-preview) - Score: 35.98
- [o1-pro](/models/o1-pro) - Score: 35.72
- [GPT-OSS 20B](/models/gpt-oss-20b) - Score: 33.32
- [GPT-4o](/models/gpt-4o) - Score: 29.92
- [GPT-4.1 mini](/models/gpt-4-1-mini) - Score: 29.27
- [GPT-4o mini](/models/gpt-4o-mini) - Score: 26.8
- [GPT-4.1 nano](/models/gpt-4-1-nano) - Score: 24.53
- [GPT-4 Turbo](/models/gpt-4-turbo) - Score: 21.54
- [GPT-4.5](/models/gpt-4-5) - Score: not computed
- [GPT-5.1-Codex-Max](/models/gpt-5-1-codex-max) - Score: not computed
- [o4-mini](/models/o4-mini) - Score: not computed
- [o4-mini (high)](/models/o4-mini-high) - Score: not computed
- [o1-mini](/models/o1-mini) - Score: not computed
- [GPT-5.2 Instant](/models/gpt-5-2-instant) - Score: not computed
- [GPT-5.3 Instant](/models/gpt-5-3-instant) - Score: not computed
- [GPT-5 (high)](/models/gpt-5-high) - Score: not computed
- [GPT-5.3-Codex-Spark](/models/gpt-5-3-codex-spark) - Score: not computed
- [GPT-5.6 Luna (low)](/models/gpt-5-6-luna-low) - Score: not computed
- [GPT-6 Luna (low)](/models/gpt-6-luna-low) - Score: not computed
- [GPT-6 Luna (medium)](/models/gpt-6-luna-medium) - Score: not computed
- [GPT-Live-1](/models/gpt-live-1) - Score: not computed
- [GPT-5.6 Cyber](/models/gpt-5-6-cyber) - Score: not computed
- [GPT Live Transcribe](/models/gpt-live-transcribe) - Score: not computed
- [GPT Transcribe](/models/gpt-transcribe) - Score: not computed
- [GPT Realtime 2.1](/models/gpt-realtime-2-1) - Score: not computed
- [GPT Realtime 2.1 Mini](/models/gpt-realtime-2-1-mini) - Score: not computed
- [GPT Audio 1.5](/models/gpt-audio-1-5) - Score: not computed
- [GPT Realtime 1.5](/models/gpt-realtime-1-5) - Score: not computed
- [GPT Realtime mini](/models/gpt-realtime-mini) - Score: not computed
- [GPT Realtime](/models/gpt-realtime) - Score: not computed
- [GPT-4o mini TTS](/models/gpt-4o-mini-tts) - Score: not computed
- [GPT-4o mini Audio](/models/gpt-4o-mini-audio) - Score: not computed
- [GPT-4o Audio](/models/gpt-4o-audio) - Score: not computed
- [Whisper large-v3-turbo](/models/whisper-large-v3-turbo) - Score: not computed
- [GPT Realtime 2](/models/gpt-realtime-2) - Score: not computed