Meta open-sources Muse Spark Open to mixed reception: 'trash' license, nerfed benchmark scores
gabriberton · x · 2026-10-17
Meta released the weights of Muse Spark Open, sparking a community fight:
- Artificial Analysis index: Muse Spark 1.3 (max) tops the chart at 48, ahead of MiniMax M2.6 (46), GLM 5.3 (45), Kimi K3 (44) — but the open-base Muse Spark Open scores just 39. tiny corp's George Hotz says Meta 'nerfed' its open release and missed the top open-source spot
- Nathan Lambert: "The Meta Muse Open License is trash," noting only inference providers will actually comply
- Poster gabriberton defends Meta: best non-Chinese open model yet — 'a bad not-so-open model beats no model at all'
A textbook clash over open-source sincerity, license restrictions, and benchmark strategy.
More from Models
- AndrewDai launches Glimpse, a 'visual thinking model' claiming frontier-level visual reasoning vs Astra — AndrewDai · 2026-10-17
- SGLang powers Reflection AI's new Beam model in production with day-0 support — ying11231 · 2026-10-17
- Unverified: Qwen 4 lineup to heavily use recursive self-improvement, index up 40→45 — solyarisoftware · 2026-10-17
- Pipecat's open PhoneLLM Max Alpha 2 tops voice-agent bench at 83%, beats Sonnet 5.5 — solyarisoftware · 2026-10-17
- NVIDIA releases LiLiCorr speculative decoding draft for Nemotron 3.5 Bolt to speed up 74B serving — _akhaliq · 2026-10-17
- US labs distill from the same APIs too, just via data vendors — peterjliu · 2026-10-17