Frontier Log

Benchmark · InactiveSuperseded by OSWorld 2.1

OSWorld 2.0, no release named

Created by XLANG Lab, 28 Jun 2026 · Higher is better · Website

020406080Jul 202670.6%Opus 562.6%GPT-5.6 Sol45.6%GPT-5.6 Luna020406080Jul 202670.6%Opus 562.6%GPT-5.6 Sol45.6%GPT-5.6 Luna

Results

ModelScoreSource
Claude Opus 524 Jul 2026 · Anthropic70.6%Introducing Claude Opus 5Checked A person checked this number against its source on 18 Oct 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.Source image
GPT-5.6 Sol9 Jul 2026 · OpenAI62.6%GPT-5.6: Frontier intelligence that scales with your ambitionChecked A person checked this number against its source on 18 Oct 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.Our capture
GPT-5.6 Terra9 Jul 2026 · OpenAI50.2%GPT-5.6: Frontier intelligence that scales with your ambitionCorroborated The same number is printed in 3 other lab documents: 1 2 3Our capture
GPT-5.6 Luna9 Jul 2026 · OpenAI45.6%GPT-5.6: Frontier intelligence that scales with your ambitionChecked A person checked this number against its source on 26 Oct 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.Our capture

All benchmarks