Hands-on: Meta's Muse Spark Open loses to smaller Chinese flash models
karminski3 · x · 2026-10-17
A hands-on speed test of Meta's newly open-sourced Muse Spark Open concludes it can only compete among Chinese flash-tier models, not SOTA ones.
- Its only bright spot is complex algorithm implementation; 3D rendering/modeling and spatial understanding are weak.
- With 42B activated parameters, local deployment makes little sense — DeepSeek-v4.1-flash prefill activates only 8B and outperforms it.
- Notable hallucination quirk: among 70+ models tested, it's the only one that drew the floppy disk write-protect slider in red when the prompt specified a blue line sketch.
More from Models
- Meta drops 816B open-weight MoE, Anthropic hits $16.8B Q3, 8 states sue OpenAI — rohanpaul_ai · 2026-10-17
- Andriy Burkov: the only gap between open-weight and frontier models is data quality — burkov · 2026-10-17
- NVIDIA open-sources midtraining recipe: 500B tokens lift Nemotron agentic benchmark scores — denisyarats · 2026-10-17
- Liquid AI ships d1-ultra decision model, tops Artificial Analysis Decision Index at 65 — DeryaTR_ · 2026-10-17
- Quick take: Claude 5.5 Haiku trails 6.1 Luna, which roughly matches Gemini 3.8 Flash — burkov · 2026-10-17
- Anthropic launches Claude Haiku 5.5: 75% cheaper, 39.2% on Terminal-Bench vs GPT-6 Luna's 16.4% — dl_weekly · 2026-10-17