Using Muse Spark Open on Together AI

Implementation guide · Muse · AI at Meta

ServerlessOpen Weights

Together AI exposes Muse Spark Open through model ID meta-models/Muse-Spark-Open-816B-A42B. Use the setup steps, sourced pricing, capabilities, and official provider links below to validate this route before deployment.

Last refreshed 2026-10-15. Next refresh: weekly.

Quick Start

  1. 1
    Create an account at Together AI and generate an API key.
  2. 2
    Use the Together AI SDK or REST API to call meta-models/Muse-Spark-Open-816B-A42B — see the documentation for request format.
  3. 3
    You'll be billed $1.25/1M input, $4.25/1M output tokens. See full pricing.

Code Examples

Install
pip install together
API key
TOGETHER_API_KEY
Model ID
meta-models/Muse-Spark-Open-816B-A42B

Together uses "organization/model-name" format, e.g. "meta-llama/Llama-4-Scout-17B-16E-Instruct" or "Qwen/QwQ-32B". See the Together model catalog for the exact ID.

from together import Together

client = Together()  # reads TOGETHER_API_KEY from env
response = client.chat.completions.create(
    model="meta-models/Muse-Spark-Open-816B-A42B",
    messages=[{"role": "user", "content": "Hello"}]
)
print(response.choices[0].message.content)

Pricing on Together AI

TypePrice (per 1M)
Input tokens$1.25
Output tokens$4.25

Capabilities

VisionMultimodalReasoningJSON / Tool useAudio

About Muse Spark Open

Muse Spark Open is the open-weight release of Meta Superintelligence Labs' Muse Spark, an 816-billion-parameter mixture-of-experts model that activates about 42 billion parameters per token. Released October 15, 2026 under the Muse Model License, it is built for coding and agentic work, takes text, image, video and audio input, returns text, and has a 1M-token context window. Weights are on Hugging Face in BF16 and NVFP4.

Model Specs

Released2026-10-15
Parameters816B (42B activated)
Context1M
ArchitectureMixture of Experts

Provider

Together AI

San Francisco, California, United States