Frontier Log

Benchmark

OSWorld 2.0

020406080Oct 202672.6%GPT-6 Astra69.2%Gemini 4 Argon020406080Oct 202672.6%GPT-6 Astra69.2%Gemini 4 Argon

Results

ModelScoreSourceHuman-verified
Gemini 4 Argon30 Sep 2026 · Google69.2%Gemini 4 Argon: our next era of frontier intelligenceSource imageYes
GPT-6 Astra3 Sep 2026 · OpenAI72.6%GPT-6 Astra: A new generation of intelligencetask release v2026.08.08Yes

Benchmark

huggingface.coHigher is better

All benchmarks