---
id: "meta/muse-spark-open"
also: ["meta/muse-spark-open-20261015", "meta-models/Muse-Spark-Open-816B-A42B", "hf/meta-models/Muse-Spark-Open-816B-A42B", "requesty/meta/muse-spark-open", "vercel/meta/muse-spark-open"]
cheapest_list_usd_per_m: "1.25 in, 4.25 out, at baseten on hf"
cheapest_effective_usd_per_m: "0.2724 in, 4.25 out, at together on requesty"
context: 1048576
max_output: 262144
input: ["text", "image", "video", "file", "audio"]
output: ["text"]
released: "2026-10-15"
reasoning: ["xhigh", "high", "medium", "low"]
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "include_reasoning", "logit_bias", "logprobs", "max_tokens", "min_p", "presence_penalty", "reasoning", "reasoning_effort", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 13
gateways: "openrouter (5), vercel (5), requesty (2), hf (1)"
no_training: 12
trains: 0
retains_prompts: 0
intelligence: 24.9000
---
# Meta: Muse Spark Open

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.