Meta's Muse Spark 1.3 scores 68 on coding agent index, near Claude Opus 5 at fraction of cost
ArtificialAnlys · x · 2026-09-03
Artificial Analysis benchmarks Meta's Muse Spark 1.3: the max variant (limited preview) scores 68 on the Coding Agent Index in Muse Code, tied with Claude Opus 5 (xhigh) and ahead of Claude Fable 5 (67) and GPT-5.6 Sol (65). The available xhigh variant scores 64 and costs just $1.72 per task — the cheapest above 60 and 1/5 of Opus 5's $8.17. Max averages 14M tokens per task, a third fewer than Opus 5 at the same score.
Related event: Meta Launches Muse Spark 1.3: Near-Frontier Performance at ~1/8 the Price(53 posts)→
More from Models
- Opus 5.5 tops OpenRouter spending in 20 of 29 task categories — mmurph · 2026-10-21
- Meta team builds open-source 30B LLM from scratch in months, hails 'second era' — TimDarcet · 2026-10-21
- DFlash drafter for Muse Glimmer-30B: 5 layers, blockwise 16-token spec decoding — TimDarcet · 2026-10-21
- EvoPress quantization shrinks 30B model from 56GB to 19GB at near-bf16 accuracy — TimDarcet · 2026-10-21
- Radar chart: Muse Glimmer-30B benchmarked against Gemma4-31B and Qwen3.6-27B — TimDarcet · 2026-10-21
- Privacy evals: Muse Glimmer-30B leaks less than Qwen3.6-27B on CIMemories — TimDarcet · 2026-10-21