All models

Meta

Muse Spark Open balancedruns locally

meta/muse-spark-open

Muse Spark Open is a coding-optimized open-weight model. Trained specifically for agentic coding workflows, it delivers higher first-attempt accuracy, reliable tool calling, and deep codebase understanding across the repos developers...

Input / 1M

$1.25

USD per million tokens

Output / 1M

$4.25

USD per million tokens

Est. cost / token

$0.000002

$2.00 per 1M, 3:1 mix

Context window

1.05M

max output 944K

Major features

VisionOpen weightsCachingReasoningJSON schemaTools

Task suitability (heuristic 1–5, not a benchmark)

  • coding4
  • reasoning4
  • math4
  • summarization4
  • extraction4
  • classification4
  • chat4
  • creative4
  • translation4
  • vision4
  • long context5
  • agentic4

Run it locally

Will it run on my machine? →
QuantizationQualityMemory needed (8K ctx)
Q2_Kpoor370 GB
Q3_K_Mfair430 GB
Q4_K_Mrecommendedgood536 GB
Q5_K_Mvery good622 GB
Q6_Kexcellent716 GB
Q8_0near-lossless925 GB
F16lossless1734 GB

Estimates. Mixture-of-experts: memory covers all parameters; only the active ones are used per token, so it runs faster than its size suggests.

Price history

RecordedInputOutputCachedSource
2026-11-05 03:20$1.25$4.25$0.15openrouter

A row is added whenever a sync detects a price change. Vendor pricing source

Pricing (USD per 1M tokens)

Input
$1.25
Output
$4.25
Cached input
$0.15
Blended (3:1)
$2.00

Details

Company
Meta
Version
—
Family
Muse
Released
2026-10-15
Parameters
822B (42B active)
Max output
944K
Open weights
Yes
Inputs
audio, file, image, text, video
Last synced
32 min ago