Skip to content
LLMs
Meta logo

Meta: Muse Spark Open

by Meta

Meta: Muse Spark Open is a large language model from Meta. It costs $1.25 per million input tokens and $4.25 per million output tokens. Its context window is 1.0M tokens.

ReasoningTool callingStructured outputPrompt cachingOpen weightsimage inputvideo inputfile inputaudio input
Input / 1M tokens
$1.25
Output / 1M tokens
$4.25
Cached input / 1M
$0.150

On repeated prefixes

Context window
1.0M

tokens

Who serves it cheapest

6 hosts serve Muse Spark Open. Same weights, same API - the price difference is pure margin and routing.

Providers serving Meta: Muse Spark Open, cheapest first
ProviderInput / 1MOutput / 1MContextThroughputUptime 24h
MakoraCheapestfp4$1.25$4.251.0M-100.0%
Modal$1.25$4.251M-100.0%
Together$1.25$4.251.0M-100.0%
Nebiusfp4$1.25$4.251.0M-89.8%
Fireworks$1.25$4.251.0M-98.4%
BaseTenfp4$1.25$4.251.0M-100.0%

The spread between Makora and BaseTen is about the same for identical weights. Quantization and context limits differ, so check both columns before switching.

About 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...

Specifications

Meta: Muse Spark Open specifications
Model IDmeta/muse-spark-open
ProviderMeta
Context window1.0M tokens
Max output944K tokens
Input modalitiestext, image, video, file, audio
Output modalitiestext
Knowledge cutoff-
Open weightsYes - meta-models/Muse-Spark-Open-816B-A42B
ReleasedOctober 15, 2026

Cheaper alternatives

Models that cost less than Muse Spark Open while keeping at least half its context window and every input modality it supports.

Frequently asked

How much does Meta: Muse Spark Open cost?

$1.25 per million input tokens and $4.25 per million output tokens. Cached input reads cost $0.150 per million tokens.

What is the context window of Meta: Muse Spark Open?

1.0M tokens, with up to 944K tokens of output per request.

Which provider serves Meta: Muse Spark Open cheapest?

Makora at $1.25 per million input tokens - about the same cheaper than BaseTen, the most expensive of the 6 hosts serving it.