---
id: "meta/muse-spark-1.1"
also: ["meta/muse-spark-1.1-20260709", "vercel/meta/muse-spark-1.1"]
card_usd_per_m: "1.25 in, 4.25 out, at meta"
cheapest_list_usd_per_m: "1.25 in, 4.25 out, at meta on openrouter"
cheapest_effective_usd_per_m: "0.5343 in, 4.25 out, at meta on vercel"
context: 1048576
max_output: 943718
input: ["text", "image", "video", "file"]
output: ["text"]
released: "2026-07-16"
reasoning: ["xhigh", "high", "medium", "low", "minimal"]
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["include_reasoning", "max_tokens", "reasoning", "reasoning_effort", "repetition_penalty", "response_format", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_p"]
endpoints: 2
gateways: "openrouter (1), vercel (1)"
no_training: 2
trains: 0
retains_prompts: 1
coding: 71.3000
agentic: 25.8000
---
# Meta: Muse Spark 1.1

Muse 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.