Muse Spark 1.2 logo

Muse Spark 1.2

Meta · Sorti août 2026

Conditionnel

Meta's coding-focused model update, co-trained with the Muse Code harness — 82.9% on Terminal-Bench 2.1 and 59.3% on DeepSWE 1.1, second only to Claude Opus 5 on the former and trailing both Opus 5 and GPT-5.6 Terra on the latter. Standard pricing unchanged from 1.1 at $1.25 input / $4.25 output per 1M tokens; contributor tier drops to $0.10/$0.20 in exchange for training-data rights. All published benchmarks remain vendor-run with no independent reproduction. Best for cost-sensitive coding teams willing to test a model explicitly optimized for its own harness.

Est-ce fait pour vous ?

Recommandé pour

  • Cost-sensitive coding agent workloads — $1.25/$4.25 standard pricing undercuts Claude, GPT, and Gemini flagships by 3-5x
  • Muse Code agent integration — co-trained with the harness for best-in-tool performance
  • Long-horizon autonomous coding — 1,000+ tool calls over 24 hours, event-log crash recovery, sustained optimization past initial exploration
  • Multimodal repo understanding — accepts text, image, video, and PDF inputs for coding context

Déconseillé pour

  • Production coding accuracy without independent validation — all published benchmarks are vendor-run in Meta's own framework; no third-party reproduction exists yet
  • Non-coding or general agentic tasks — coding-focused checkpoint; reasoning, knowledge, math, and multilingual benchmarks remain unpublished
  • Open-weight or self-hosting requirements — closed weights, API-only; no Hugging Face weights, no fine-tuning
  • High-concurrency production on contributor tier — 60 RPM cap vs 3,000 standard; training-data grant on all contributor traffic

Performances par tâche

Code generation

Good

Second on Terminal-Bench 2.1 (3.8 pts behind Opus 5) and Meta's internal coding bench (8.8 pts behind); third on DeepSWE 1.1 behind Opus 5 and GPT-5.6 Terra

Agentic coding

Very Good

Persistent background agents + worktree isolation; 24-hour kernel optimization demo with 1,000+ tool calls; co-trained with Muse Code harness

Long-horizon tasks

Very Good

Sustained improvement over 24 hours on GPU kernel optimization; event-log replay on crash prevents lost work and re-prompting

General reasoning

Fair

Coding-focused checkpoint; reasoning, knowledge, math, and multilingual benchmark categories remain unpublished — cannot assess non-coding strength

Tarifs

Entrée

$1.25 / 1M tokens

Sortie

$4.25 / 1M tokens

Contexte

1M tokens

Voir tous les tarifs

Tests de performance

TestScoreSource
Terminal-Bench 2.182.9% Source
DeepSWE 1.159.3% Source
Meta Internal Coding Bench70.6% Source
BenchLM Aggregate60.3/100 (#49 of 216) Source

Aucun changement de verdict pour l’instant

L’horloge tourne dès le premier jour : les changements apparaîtront ici à mesure que notre verdict évolue.

Journal de vérification

Aucune vérification pour le moment

Nous n’avons pas encore enregistré de vérification pour cette entrée. Dès qu’une vérification s’exécute, son historique apparaît ici.

Comment nous évaluons