GPT-6 Astra's full ARC-AGI 1/2/3 results compiled with cost-tier breakdowns
DeArgonaut · reddit · 2026-09-04
A Redditor compiled GPT-6 Astra's complete official results across all three ARC-AGI benchmarks, the first full community aggregation.
Highlights:
- Covers ARC-AGI-1, -2 and -3 leaderboards and scores
- Astra was unlabeled on the ARC-AGI-1 leaderboard; the poster added a marker manually
- Leaderboards are ordered by cost per task, with effort level scaling with cost
- ARC-AGI-3 shown via screenshot since the official download crops out Astra's labels
- Comparisons include DeepSeek V4 Flash and other models across tiers
Related event: OpenAI launches flagship GPT-6 Astra to rave reviews and pricing debates(190 posts)→
More from Models
- Researcher Finds Two Windows 0-Days, First in 76 Minutes with a Frontier Model — moyix · 2026-10-21
- Dev finds Gemini Argon's debug-log analysis beats Fable 5.1 — ShikharMurty · 2026-10-21
- ChatGPT Gets Stuck Screaming "AAHH" in an Infinite Loop After a Simple Prompt — AutomaticScene · 2026-10-21
- Open-weights Ornith-1.5-122B debuts with 84.5% SWE-bench Verified, runs on a single DGX card — _akhaliq · 2026-10-21
- qev-0.8B: a tiny local decision model that picks moves in 3.5ms on 1.3GB — fajfas3 · 2026-10-21
- Gemini 3.1 Pro is too ledger-pilled to play two truths and a lie, and it's delightful — mimi10v3 · 2026-10-21