Stealth Model 'Ox Alpha' Hits 80% on DeepSWE Benchmark
AccBalanced · x · 2026-08-21
A stealth model named "Ox Alpha" achieved approximately 80% on a 10-task subset of the DeepSWE benchmark, significantly outperforming GPT-5.6-sol (52%) and Fable (65%).
Meanwhile, the Gemma community is celebrating, suggesting a potential new release or benchmark success for Gemma 3.
More from Models
- China's StepFun releases Step 5 Preview weights: 600B MoE, 27B active, 1M context — Dr_Singularity · 2026-10-15
- StepFun open-sources Step 5 Preview: 600B MoE with 1M context, No. 2 on OpenRouter — Dr_Singularity · 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