GPT-6 Astra

OpenAI · openai/gpt-6-astra

ProviderOpenAI
Familygpt-astra
Typellm-reasoning
Statusactive
Released2026-09-04
Updated2026-09-04

Capabilities

Chain Of Thought · tier-2 Function Calling · tier-2

Verified benchmark evidence

Each row was checked against its source by a reviewer, and carries the model identifier as actually evaluated — which is not always the same as this card's.
BenchmarkEvaluated asScoreEvidence dateSource kind
aa_briefcaseGPT-6 Astra (max)53.0 normalized Elo percent2026-09-04 publishedindependent evaluatorsource
aa_lcrGPT-6 Astra (max)81.0%2026-09-04 publishedindependent evaluatorsource
automationbench_aaGPT-6 Astra (max)68.5%2026-09-07 publishedindependent evaluatorsource
critptGPT-6 Astra (max)32.0%2026-09-04 publishedindependent evaluatorsource
gdp_pdf_aaGPT-6 Astra (max)33.0%2026-09-04 publishedindependent evaluatorsource
gdpval_aaGPT-6 Astra (max)54.0 normalized Elo percent2026-09-04 publishedindependent evaluatorsource
scicodeGPT-6 Astra (max)56.0%2026-09-04 publishedindependent evaluatorsource

Reported benchmark scores

This card reports no benchmark scores yet.

BenchmarkCatalogue standingScore

Data

This card as JSON · See it in the graph · Edit on GitHub