Ox-alpha DeepSWE benchmark reveals 58.4% pass rate, debunking 80% rumors
apples_jimmy · x · 2026-08-22
A full end-to-end run of the DeepSWE benchmark on ox-alpha revealed a pass rate of 58.4% (66/113 tasks), debunking rumors of an 80% score. This performance lands the model nearly identical to Claude Opus 4.8 (59%). The test noted that 9.7% of failures were due to tool-call format errors rather than reasoning shortcomings.
Related event: Ox Alpha scores ~63% on full DeepSWE; earlier 80% was subset artifact(6 posts)→
More from Models
- Unredacted Anthropic filing exposes internal data-quality debate months before Opus 3 cutoff — zacharynado · 2026-10-15
- Google says teams across the company are using Gemini 4 Argon on its hardest engineering problems — CurieuxExplorer · 2026-10-15
- Perplexity's new embedding models: index with a 9B model, search with 0.6B — antoine_chaffin · 2026-10-15
- 797GB of n-gram counts in CPU RAM let 1B models match 57% more compute — PangWeiKoh · 2026-10-15
- Leak: Moonshot's Kimi flagship 'x4' done with pre-training, K3.1 in internal testing — lyraxana · 2026-10-15
- Unsloth ships GGUF quants for Nemotron 3.5: Bolt-74B runs in 24GB RAM — danielhanchen · 2026-10-15