GPT-6 Astra High

Released2026-09-04
Price per 1M$10.00 in · $50.00 out
Overall score64.2 · rank 28
Context window1,050,000 tokens
ProviderOpenAI
WeightsProprietary
ConfigurationHigh

Scores

DimensionScoreRankBenchmarks
Overall: Balanced 64.2 28 12
Overall: Coding 62.1 61 9
Overall: Reasoning 68.8 10 6
Overall: Agentic work 63.4 35 9
Overall: Frontend 60.5 63 3
Coding 60.5 80 3
Agents 63.4 18 1
Reasoning 68.4 15 5
Math 69.4 4 1
General 63.5 26 2
Value vs price+5.5124

Buzz

Posts on XAbout 195 a day
Mood12% positive, 24% negative
Hacker News29 mentions this week
Checked2026-11-11

Evidence

Coding 60.5 ±2.6

BenchmarkResultSourceDate
MirrorCode 46.7% Epoch AI 2026-08-30
DeepSWE 73.2% Epoch AI 2026-11-11
SciCode 55.4% Epoch AI 2026-11-11

Agents 63.4 ±0.9

BenchmarkResultSourceDate
Terminal-Bench 4 60.0% Terminal-Bench 2026-09-03

Reasoning 68.4 ±1.8

BenchmarkResultSourceDate
Extended NYT Connections 97.7% Lech Mazur 2026-11-10
ARC-AGI-2 92.1% Epoch AI 2026-11-11
ARC-AGI-2 (ARC Prize verified) 92.1% ARC Prize 2026-11-11
CritPt 28.9% Epoch AI 2026-11-11
ARC-AGI-1 (ARC Prize verified) 98.5% ARC Prize 2026-11-11
ARC-AGI-1 98.5% Epoch AI 2026-11-11
ARC-AGI-3 100.0% ARC Prize 2026-11-11

Math 69.4 ±1.8

BenchmarkResultSourceDate
FrontierMath Tier 4 97.6% Epoch AI 2026-08-30

General 63.5 ±2.8

BenchmarkResultSourceDate
LLM Debate Benchmark 1,604 Lech Mazur 2026-10-26
Creative Story-Writing (Lech Mazur) 91.0% Lech Mazur 2026-10-23

Results not used in the score

BenchmarkResultSourceWhy not used
ARC-AGI-3 54.8% ARC Prize Lower duplicate run on the same benchmark

Names matched to this model

gpt-6-astra-high

Compare GPT-6 Astra High with other models