Skip to content
English

Switch theme

OpenAI

GPT-6 Astra benchmark results

9 results on 5 benchmarks from 2 sources.

Closed Released: Sep 2, 2026 Updated Oct 22, 2026 API prices & specs →

Also listed as: gpt-6-astra_max · GPT-6 Astra (max) · gpt-6-astra_unknown · GPT-6 Astra (unknown thinking) · GPT 6 Astra · GPT-6 Astra (XHigh) · GPT-6 Astra (High) · GPT-6 Astra (Max) · GPT-6 Astra (Medium)

All benchmark results

Benchmarks Score Rank Setting Source type Date
GPQA Diamond 95.8% 2 Epoch AI run, accuracy; effort: max Independent run Aug 30, 2026
AIME (OTIS Mock AIME 2024–2025) 100.0% 1 Epoch AI run, OTIS Mock AIME 2024–2025; effort: max Independent run Aug 30, 2026
Humanity's Last Exam 54.8% 1 Compiled by Epoch AI; effort: unknown thinking Third-party compilation via Epoch AI Sep 3, 2026
ARC-AGI-1 98.5% 3 Semi-private set; $0.35/task; CoT“GPT-6 Astra (XHigh)” Official leaderboard Oct 22, 2026
ARC-AGI-1 98.5% 4 Semi-private set; $0.28/task; CoT“GPT-6 Astra (High)” Official leaderboard Oct 22, 2026
ARC-AGI-2 95.0% 1 Semi-private set; $1.12/task; CoT“GPT-6 Astra (Max)” Official leaderboard Oct 22, 2026
ARC-AGI-2 93.3% 3 Semi-private set; $0.83/task; CoT“GPT-6 Astra (XHigh)” Official leaderboard Oct 22, 2026
ARC-AGI-2 92.1% 7 Semi-private set; $0.67/task; CoT“GPT-6 Astra (High)” Official leaderboard Oct 22, 2026
ARC-AGI-2 92.1% 8 Semi-private set; $0.48/task; CoT“GPT-6 Astra (Medium)” Official leaderboard Oct 22, 2026
← AI Model Benchmarks

FAQ

Where do these GPT-6 Astra scores come from?

From Epoch AI Benchmarking Hub and ARC Prize leaderboard. Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.

Data updated: Benchmarks(UTC+8)