US open-weights models jump in a week: Meta Muse Spark Open hits 38.9, Reflection's Beam 1.0 lands at 30.3
ArtificialAnlys · x · 2026-10-20
Artificial Analysis recaps a fast week for US open-weight models: NVIDIA's Nemotron 3.5 Super VL scored 26.5 on the Intelligence Index with 120B parameters, pushing Inkling Small off the Pareto frontier; Meta's Muse Spark Open scored 38.9, over 12 points ahead and the highest yet for a US open-weights model; and Reflection's first model, Beam 1.0 (501B parameters), scored 30.3, joining the frontier between the two. Open-weight models advanced on both intelligence and parameter efficiency in days.
More from Models
- DeepSeek hands inference providers 90%+ margins, pressuring Moonshot and Zhipu hosting deals — teortaxesTex · 2026-10-20
- HelixWorld world model tested: 70GB VRAM per BF16 GPU run, consumer cards need not apply — cocktailpeanut · 2026-10-20
- Code-only Paris at golden hour: Claude's render embarrasses ChatGPT in same-prompt test — EXM7777 · 2026-10-20
- Prompting can't fix mode collapse: making agents sound human requires midtraining — CShorten30 · 2026-10-20
- Tess-4.5-27B released: open fine-tune beats its Qwen base using 30% fewer reasoning tokens — udmrzn · 2026-10-20
- Same Agentic Task: Gemini 4 Argon Costs $1.13 vs Claude Opus 5.5's $114.34 — DynamicWebPaige · 2026-10-20