All AI models · LLM benchmarks · Methodology

Meta

Muse Spark 1.3

Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It improves long-horizon agent collaboration, instruction following, and coding efficiency relative to Muse Spark 1.2.

Model facts

Context window
1.05M tokens
Maximum output
944K tokens
Cheapest paid input
$1.25 per 1M tokens
Cheapest paid output
$4.25 per 1M tokens
Open weights
no
Providers
8

Capabilities and modalities

reasoning, tool calling, structured output, attachments, temperature control, text, image, audio, video, pdf.

Artificial Analysis benchmarks

Compare this model on the full LLM benchmark leaderboard.

ConfigurationAAICodingMathOutput tokens/s
max62.176.3not measurednot measured
xhigh60.876.5not measured110.8

Observed price history

  • 2026-09-03: $1.25 input / $4.25 output per 1M tokens

API providers

  • DevPass (LLM Gateway) — model id muse-spark-1.3; $1.25 input / $4.25 output per 1M tokens; provider documentation
  • EmpirioLabs AI — model id muse-spark-1-3; $1.25 input / $4.25 output per 1M tokens; provider documentation
  • Kilo Gateway — model id meta/muse-spark-1.3; $1.25 input / $4.25 output per 1M tokens; provider documentation
  • LLM Gateway (Meta) — model id meta/muse-spark-1.3; $1.25 input / $4.25 output per 1M tokens; provider documentation
  • Meta — model id muse-spark-1.3; $1.25 input / $4.25 output per 1M tokens; provider documentation
  • NanoGPT — model id meta/muse-spark-1.3; $1.25 input / $4.25 output per 1M tokens; provider documentation
  • OpenRouter — model id meta/muse-spark-1.3; $1.25 input / $4.25 output per 1M tokens; provider documentation
  • Vercel AI Gateway — model id meta/muse-spark-1.3; $1.25 input / $4.25 output per 1M tokens; provider documentation