Muse Spark

Muse Spark is Meta's model for agentic and coding work at scale, such as multi-step tool loops, software engineering assistants, sub-agent tasks, and long-context reasoning. It's the lower-cost option for high-volume, latency-sensitive work, and on Model API it also offers a discounted Contributor mode.

At a glance

Standard mode model IDsmuse-spark-1.3, muse-spark-1.2, muse-spark-1.1
Contributor-mode variantsmuse-spark-1.3-contributor, muse-spark-1.2-contributor
InputText, image, video, audio, and PDF
OutputText
Context window1,048,576 tokens
Reasoning effort"minimal" to "max" on muse-spark-1.3; see Versions
EndpointsResponses, Chat Completions, Messages
PricingPay-as-you-go; see Muse Spark pricing
Rate limitsSee rate limits

Build with Muse Spark

Use Muse Spark for high-volume, latency-sensitive, or cost-sensitive work:

  • Sub-agent tasks: run the many short, focused tasks that a main agent hands out.
  • Tight tool loops: drive multi-step tool calls where latency per turn adds up.
  • Classification: process large volumes of requests at a lower cost per token.
  • Coding assistants: connect the coding agent you already use, or run Muse Code.

Muse Spark also handles image understanding, video and audio understanding, tool calling, structured output, and search grounding. For complex, long-running tasks where capability matters more than cost, see Choose a model.

Versions

Muse Spark comes in three versions. All take text, image, video, audio, and PDF input, return text, and have a 1,048,576-token context window. They differ by capability:

  • Muse Spark 1.3 (muse-spark-1.3): the latest version, tuned for agentic workflows (multi-step tool, browser, and long-horizon tasks) with improved coding over 1.2. Supports all reasoning effort levels, including the "max" level for extended reasoning (not available with muse-spark-1.3-contributor). Recommended for new work.
  • Muse Spark 1.2 (muse-spark-1.2): the previous version.
  • Muse Spark 1.1 (muse-spark-1.1): the original version.
Audio input on Muse Spark 1.3

Audio understanding in Muse Spark 1.3 is not fully supported, and response quality for requests that include audio may be degraded. For audio, use Muse Spark 1.2, or Muse Voice Transcribe for dedicated speech-to-text.

muse-spark-1.3 is the default model in most code examples in these docs. The coding agents, agent frameworks, and computer use guides use muse-wonder-1.0 for long-running agentic work.

Make your first call

Follow the quickstart to set your key and send your first request to muse-spark-1.3.

Availability

Every Muse Spark version is available on pay-as-you-go. All versions except Muse Spark 1.1 are also available in Muse Code, on pay-as-you-go and with any subscription.

VersionModel API (pay-as-you-go)Muse Code
Muse Spark 1.3 (muse-spark-1.3)Standard modePay-as-you-go and every plan
Muse Spark 1.3 Contributor (muse-spark-1.3-contributor)Contributor modePay-as-you-go and every plan
Muse Spark 1.2 (muse-spark-1.2)Standard modePay-as-you-go and every plan
Muse Spark 1.2 Contributor (muse-spark-1.2-contributor)Contributor modePay-as-you-go and every plan
Muse Spark 1.1 (muse-spark-1.1)Standard modeNot available

In Standard mode, your content — the prompts, code, files, and context you send, plus the model responses you get back — is not used to train Meta models. In Contributor mode, your content may be used to train future Meta models, in exchange for a lower price. On pay-as-you-go, switch between the two by choosing the model ID: for example, muse-spark-1.3 for Standard mode or muse-spark-1.3-contributor for Contributor mode. Muse Spark 1.1 is available in Standard mode only. See Pricing and rate limits for rates.

Next steps

  • Weighing Muse Spark against Muse Wonder? Choose a model compares them workload by workload.
  • Get schema-guaranteed JSON for your classification pipeline with structured output.
  • Run Muse Spark as the sub-agent model under a Muse Wonder main loop with an agent framework.