Benchfolio

Leaderboard / Model

GPT-6 Astra

OpenAI · Closed weights · Released Sep 3, 2026 · $10 in · $50 out per 1M tokens

87Overall · rank 3

Score by job

Coding

81 · rank 4 · 3 of 5 benchmarks

BenchmarkResultVariantz
DeepSWE74.1%gpt-6-astra_xhigh1.09
FrontierCode53.3%gpt-6-astra_max1.42
LMArena Coding1489 ratinggpt-6-astra-max-0.10

Agents

100 · rank 1 · 2 of 4 benchmarks

BenchmarkResultVariantz
APEX-Agents64.7%gpt-6-astra_unknown1.18
Remote Labor Index20.8%gpt-6-astra_unknown1.83

Reasoning

100 · rank 1 · 3 of 4 benchmarks

BenchmarkResultVariantz
GPQA Diamond95.8%gpt-6-astra_max1.70
ARC-AGI-295.0%gpt-6-astra_max1.43
Humanity’s Last Exam54.8%gpt-6-astra_unknown1.96

Math

100 · rank 1 · 4 of 5 benchmarks

BenchmarkResultVariantz
FrontierMath Tiers 1–393.7%gpt-6-astra_max1.77
FrontierMath Tier 497.6%gpt-6-astra_high2.37
OTIS Mock AIME100.0%gpt-6-astra_max1.09
ProofBench99.0%gpt-6-astra_unknown2.02

Writing

46 · rank 36 · 2 of 2 benchmarks

BenchmarkResultVariantz
LMArena Creative Writing1429 ratinggpt-6-astra-max-0.68
LMArena Instruction Following1439 ratinggpt-6-astra-max-0.94

Design

94 · rank 2 · 1 of 1 benchmarks

BenchmarkResultVariantz
LMArena WebDev1792 ratinggpt-6-astra-max2.32