Open-weight Qwen3.8 beats Claude Opus 5 on 4 of 5 enterprise personas just 15 days after launch
ryanshrout · x · 2026-09-16
On Signal65's PINNACLE enterprise benchmark, Claude Opus 5 topped the board at launch on Aug 31 — but 15 days later Alibaba's open-weight Qwen3.8-2.4T-A95B (medium reasoning effort) beats it on four of five personas and ties the fifth, making 15% fewer weighted errors on multi-step enterprise work. The best open model trailed by 1.4x at launch. Ryan Shrout: the frontier moved, but open weights closed the two-week gap — the pace any pause must contend with.
More from Models
- 'You Can Feel the Gradient Slopes': Dev Contrasts Benchmaxxed RL Models vs General Ones — theteknosaur · 2026-11-02
- LightOnOCR-3 Launches: Tops FRBench, Adds Layout Structure and Chart Extraction — IgorCarron · 2026-11-02
- NVIDIA's Kumo Is a Family of Four Tabular Foundation Models, Not Just One — pandeyparul · 2026-11-02
- SemiAnalysis: safety classifiers are underrated — Meta's Muse tops cyber benchmark at 35.2% — ThomasScialom · 2026-11-02
- DeepSeek V4.1 Flash 1017 wins over a longtime closed-source-only user: fast, cheap, reliable — _bachrc · 2026-11-02
- Local quants vs Claude on 9 real private bugs: 3-bit costs thinking time, not score — FeydRowan · 2026-11-02