muse-spark-1.3-contributor

Limited signal
Reasoning

Standard & Paid Price

Standard Price
MeterStandard Price
Input price$0.1000 / 1M tokens
Completion price$0.2000 / 1M tokens
Cache read price$0.0020 / 1M tokens
Billing details
Vendor list price comparison

Description

Meta may use the prompts and completions sent to this model to train future Meta models under its Contributor tier terms, so do not send sensitive or confidential data. Meta Muse Spark 1.3 (Contributor tier) is the latest Muse Spark version, tuned for agentic workflows and coding. It is a reasoning model: it always reasons before it answers, and reasoning cannot be turned off. Set reasoning_effort to minimal, low, medium, high, or xhigh; none and max are not supported on this tier, and when reasoning_effort is omitted the model chooses the level. Reasoning tokens are billed as output, and cached input is billed at the cached-input rate. Web search is not available, and requests that include web search or any tool type other than function are refused. Send requests to /v1/chat/completions.

Capabilities

Input modalities
text
Output modalities
text
Tasks
chat, reasoning

API endpoints

  • POST /v1/chat/completions (openai)