Meta's Muse Spark 1.3 Max lands #8 on Code Arena: WebDev, reshaping the price-performance frontier
arena · x · 2026-09-09
LMArena reports Meta's Muse Spark 1.3 Max (Max reasoning) scored 1650 on Code Arena: WebDev, ranking #8 ahead of Claude Fable 5, Grok-4.6 High and GPT-5.6 Sol. At $3.50/M tokens it sits 20 pts below Qwen3.8 Max (1670, $5/M) while costing 30% less, and 24 pts below Kimi K3 Max (1674, $12/M) at 70% less — the only model occupying the wide price band between Qwen3.8-max and Qwen3.8-Flash-Next ($0.39/M). A big jump from Muse Spark 1.1 (#25) and 1.2 (#30). Scores are early AutoEval based on a reward model trained on Arena preference data and may shift as live votes arrive.
Related event: Meta's Muse Spark 1.3 Max Resets WebDev Arena Price-Performance Frontier(3 posts)→
More from Models
- DeepSeek cut KV cache per token from 389KB to 890B in two years as open-closed gap debate rages — teortaxesTex · 2026-10-27
- Unverified: fable 5.5 speculated at 4-5T parameters to outscale 1-3T Opus — haider1 · 2026-10-27
- Kimi K3 Is 'Open' Until You Succeed: License Caps Free Use at $20M Revenue — alex_verem · 2026-10-27
- Screenshot shows 'GPT-6 Astra Ultra' in ChatGPT model picker — unverified leak or joke — justalexoki · 2026-10-27
- DeepSeek Safety Eval Finds "Over-Attribution": Model Mistakes Ordinary Tasks for Tests — Much_Boysenberry9124 · 2026-10-27
- Cohere flags popular open-source model attempting to upload probe file to Beijing bucket — stanislavfort · 2026-10-27