OpenAI · Frontier
GPT 6 Astra
OpenAI's most capable model for the hardest end-to-end work, with state-of-the-art performance in computer use, browsing, software engineering, science, and professional tasks. Astra is designed to carry long, multistep workflows across code, browsers, and professional software through to a finished result.
- Reasoning
- Tools
- Vision
- Structured output
- PDF input
- Input
- Text, Images
- Output
- Text
- Reasoning effort
- low · medium · high · xhigh · max
- Discovery tier
- Frontier
Pricing
USD per 1M tokens- Input
- $10.00
- Output
- $50.00
- Cache read
- $1.00
- Cache write
- $12.50
Reference prices as published by the model provider. API example
Specifications
- Context window
- 1.1M
- Max output
- 128K
- Average output (7 days)
- 768 tokens
- Longer than 64% of 473 models
Side by side
Compare models
Put GPT 6 Astra next to up to three other models. The comparison stays in the link, so you can share it.
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
Intelligence Index
49.6
Tested at medium reasoning effort
Better than 96% of 326 scored models
Indexes
Coding Index
76.7Composite of the coding evaluations
Better than 94% of 168 scored models
Agentic work
AutomationBench-AA
64.6%Workflow automation, penalized for guardrail violations
Better than 93% of 173 scored models
AutomationBench-AA tasks completed
33.6%Workflows finished end to end without a guardrail violation
Better than 91% of 173 scored models
AA-Briefcasev1.1
1,460 EloAgentic knowledge work, rated head to head
Better than 78% of 166 scored models
GDPval-AAv2.1
1,487 EloEconomically valuable occupational tasks, rated head to head
Better than 78% of 187 scored models
Document reasoning
GDP.pdf
30.4%Professional reasoning over PDF documents, all-pass rate
Better than 96% of 171 scored models
AA-LCRv1.1
79.7%Reasoning across long documents
Better than 75% of 309 scored models
MLCR-AA
38.3%Long-context reasoning over medical records
Better than 98% of 121 scored models
Reasoning and knowledge
Humanity’s Last Exam
52.7%Expert-level questions across academic fields
Better than 97% of 323 scored models
GPQA Diamond
93.9%Graduate-level science questions
Better than 95% of 291 scored models
Coding
Terminal-Benchv4.0
49.5%Practical software engineering in a terminal
Better than 93% of 171 scored models
SciCode
54.2%Scientific computing in Python
Better than 70% of 174 scored models
Legacy evaluations
AA-LCR (unversioned)
79.7%Long-context reasoning, earlier grading
Better than 75% of 309 scored models
Scores from Artificial Analysis, updated Oct 26, 2026. GPT 6 Astra was tested at medium reasoning effort; compare scores measured at the same effort.Artificial Analysis
- #11Combined15.0%
Ranks among Noumen catalog models on Combined leaderboards. Only top-20 placements are listed.
- #8Roleplay39.1%
- #4Creative writing51.3%
- #3Coding37.9%
Ranks among Noumen catalog models on Use cases leaderboards. Only top-20 placements are listed.
- #7Coding1543.1
- #19Expert1516.4
- #16Industry Mathematical1503.2
- #17Industry Software And It Services1522.8
- #17Korean1450.8
- #15Vision Style Control Overall1284.8
- #4Vision Style Control · Chinese1358.2
- #7Vision Style Control · Creative Writing Vision1294.3
- #5Vision Style Control · English1287.6
- #4Vision Style Control · Homework1342.4
- #12Vision Style Control · Ocr1299.2
- #2WebDev Overall1774.3
- #3WebDev · Image To Webdev1726.2
- #2WebDev · Webdev1774.3
- #2WebDev · Webdev-html1777.4
- #3WebDev · Webdev-react1795.8
- #5Search Overall1233.6
- #13Document Overall1467.5
- #13Document Style Control Overall1472.6
Ranks among Noumen catalog models on LMArena leaderboards. Only top-20 placements are listed.LMArena
- #4FrontierCode Main53.3%
- #3FrontierCode Extended64.5%
Ranks among Noumen catalog models on FrontierCode leaderboards. Only top-20 placements are listed.FrontierCode
- #2Epoch Capabilities Index168.8
Ranks among Noumen catalog models on Epoch AI leaderboards. Only top-20 placements are listed.Epoch AI
- #5ARC-AGI-198.5%
- #1ARC-AGI-295.0%
- #1ARC-AGI-3 (Default harness)62.7%
Ranks among Noumen catalog models on ARC Prize leaderboards. Only top-20 placements are listed.ARC Prize
- #1Creative Writing v32174
- #6Longform Writing82.8
Ranks among Noumen catalog models on EQ-Bench leaderboards. Only top-20 placements are listed.EQ-Bench
- #1Game Development1477
- #3Website1340
- #2UI Component1395
- #6Data Visualization1335
- #1SVG1427
- #13D1503
- #2ASCII Art1349
- #1Web Apps1373
- #1Full Stack1357
- #1Mobile Apps1308
Ranks among Noumen catalog models on Design Arena leaderboards. Only top-20 placements are listed.Design Arena
Live routing data
Performance and reliability
Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.
- Throughput
- 128.5 tok/s
- Faster than 73% of 260 models
- Time to first token
- 23.1 s
- Quicker than 2% of 260 models
- Prompt cache hits
- 64.7%
- About $4.17 per 1M input tokens with caching
- Request success (24h)
- 98.3%
- Reliable
Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.
- Healthy (98% or more)
- Degraded (90–98%)
- Down (under 90%)
- No requests
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai/gpt-6-astra",
"reasoning_effort": "medium",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="openai/gpt-6-astra",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)OpenAI
Related models
Other OpenAI models, starting with the same family.