meta · MultimodalReleased 2026-10-15

muse-spark-open

meta/muse-spark-open id

Muse Spark Open is Meta’s open-weight reasoning model for coding and agentic workflows. Its sparse mixture-of-experts architecture has 816 billion total parameters with approximately 42 billion active per token, combining code generation, tool use, and image understanding. A one-million-token context window supports repository analysis, multi-file changes, debugging, and long-running agent tasks. This provider also documents audio and video input with text output.

Tool useReasoningStructured outputChat

Type

Use muse-spark-open

# Drop-in OpenAI-compatible client
$ import { generateText } from 'ai'
$ 
$ const { text } = await generateText({
$   model: 'meta/muse-spark-open',
$   baseURL: 'https://synapse.garden/api/v1',
$   apiKey: process.env.MG_KEY,
$   prompt: 'Why is the sky blue?',
$ })

1.0M

CONTEXT WINDOW

1.0M

MAX OUTPUT

$1.38/M

INPUT · PER M

$4.68/M

OUTPUT · PER M

PRICING

List prices, every modality.

Rate	Per million tokens · USD
Input	$1.38/M
Output	$4.68/M
Cache read	$0.165/M

Honest list pricesHow we calculate prices

Other meta models

See all 14

Model

Input

Output

Context

Type

llama-3.1-70bAn update to Meta Llama 3 70B Instruct that includes an expanded 128K context length, multilinguality and improved reasoning capabilities.TOOLS
Input
$0.792/M
Output
$0.792/M
Context
128K
llama-3.1-8bAn update to Meta Llama 3 8B Instruct that includes an expanded 128K context length, multilinguality and improved reasoning capabilities.TOOLSJSON
Input
$0.242/M
Output
$0.242/M
Context
128K
llama-3.3-70bWhere performance meets efficiency. This model supports high-performance conversational AI designed for content creation, enterprise applications, and research, offering advanced language understanding capabilities, including text summarization, classification, sentiment analysis, and code generation.TOOLS
Input
$0.792/M
Output
$0.792/M
Context
128K
llama-4-maverickAs a general purpose LLM, Llama 4 Maverick contains 17 billion active parameters, 128 experts, and 400 billion total parameters, offering high quality at a lower price compared to Llama 3.3 70B.TOOLS
Input
$0.264/M
Output
$1.07/M
Context
128K
llama-4-scoutLlama 4 Scout is the best multimodal model in the world in its class and is more powerful than our Llama 3 models, while fitting in a single H100 GPU. Additionally, Llama 4 Scout supports an industry-leading context window of up to 10M tokens.TOOLSJSON
Input
$0.187/M
Output
$0.726/M
Context
128K
muse-glimmer-30bTOOLSREASONINGJSON
Input
$0.385/M
Output
$1.65/M
Context
131.1K
muse-image-1.0Muse Image is the first image generation model from Meta Superintelligence Labs, it uses advanced reasoning to understand complex prompts, seamlessly blending multiple photos into high-quality creations you can download and share anywhere.
Input
—
Output
—
Context
—
muse-spark-1.1Muse Spark 1.1 is strongest at agentic performance, tool use, and computer use. It does well on long-running tasks with 1M token context window, can delegate execution to sub-agents running in parallel, and is trained to use computer interfaces on desktop, mobile, or browser.REASONINGTOOLSJSON
Input
$1.38/M
Output
$4.68/M
Context
1.0M
muse-spark-1.2A coding-optimized model purpose-built for agentic workflows. Improvements to code generation, debugging, and codebase understanding — with a 1M context window that handles your entire project in one session.REASONINGTOOLSJSON
Input
$1.38/M
Output
$4.68/M
Context
1.0M
muse-spark-1.2-contributorA coding-optimized model with pricing designed for builders. Same model, same capabilities — up to 95% less than Standard. Your inputs and outputs are used to train and improve Meta's AI models.REASONINGTOOLSJSON
Input
$0.110/M
Output
$0.220/M
Context
1.0M
muse-spark-1.3Muse Spark 1.3 is Meta’s multimodal reasoning model for long-horizon agentic and coding workflows. With a 1M-token context window, reliable tool calling, higher first-attempt accuracy, and native understanding of video, images, and documents, it helps developers build capable coding agents and AI development workflows with fewer unnecessary turns and cleaner output.REASONINGTOOLSJSON
Input
$1.38/M
Output
$4.68/M
Context
1.0M
muse-spark-1.3-contributorMuse Spark 1.3 Contributor is Meta’s cost-optimized API option for agentic coding workflows. It offers a 1M-token context window, dependable tool calling, and multimodal perception at $0.10 per million input tokens and $0.20 per million output tokens, with usage permitted to improve Meta’s products.REASONINGTOOLSJSON
Input
$0.110/M
Output
$0.220/M
Context
1.0M

FAQ · MUSE-SPARK-OPEN

Frequently asked

01 / 04

How do I call muse-spark-open from my code?

Use the OpenAI or Anthropic SDK and point baseURL at https://synapse.garden/api/v1. Set model: ‘meta/muse-spark-open’ and supply your Synapse Garden API key. No code changes beyond the base URL.

02 / 04

How much does muse-spark-open cost?

Input: $1.38/M per million tokens. Output: $4.68/M per million tokens. The free tier includes a million tokens every month at no cost.

03 / 04

What's the context window for muse-spark-open?

muse-spark-open supports a context window of 1.0M tokens, with a maximum output of 1.0M tokens.

04 / 04

Do I need a separate Anthropic or OpenAI account?

No. Synapse Garden is the single API surface — one key gives you OpenAI, Anthropic, Google, Meta, Mistral, DeepSeek, xAI, Cohere, and more. Billing, rate limits, and audit logs are unified.

READY

Try muse-spark-open in three minutes.

Sign up, create a key, drop our base URL into your existing client. The free tier includes a million tokens every month — no credit card.

Start free Quickstart