Disclosure: This article was written with AI assistance and reviewed by EmpirioLabs AI.
Muse Spark 1.2 is available on EmpirioLabs. This updated Meta reasoning checkpoint combines a 1,048,576-token context window with text, image, video, audio, and PDF understanding. It also supports function calling, strict structured output, built-in web search, and native Responses tools.
What Muse Spark 1.2 supports
Muse Spark 1.2 accepts text, up to 50 images, MP4 video, MP3 or WAV audio, and PDF documents. PDF processing includes text from the first 100 pages and page images from the first 50 pages, which makes the model useful for long reports, visual documents, recorded speech, and mixed-media analysis.
Reasoning is always on. Use reasoning_effort from minimal through xhigh to control how much work the model spends before answering. Reasoning traces remain private, while reasoning tokens count as output tokens and use the request's output-token budget.
How to call Muse Spark 1.2
Use the OpenAI-compatible chat completions endpoint and set model to muse-spark-1-2:
curl https://api.empiriolabs.ai/v1/chat/completions \
-H 'Authorization: Bearer $EMPIRIOLABS_API_KEY' \
-H 'Content-Type: application/json' \
-d '{
"model": "muse-spark-1-2",
"reasoning_effort": "medium",
"messages": [
{
"role": "user",
"content": "Design a safe rollout plan for a high-traffic API migration."
}
]
}'The same model is available through the Responses, Messages, and Google-compatible content-generation endpoints. The Responses endpoint preserves custom tools and deferred tool discovery with tool_search. Streaming responses include final usage data for auditable token, cache, and tool settlement.
Web search, tools, and structured output
Enable built-in web search when a response needs current information. Each executed search query is reported in usage.tool_usage and adds the catalog-listed per-search charge. Function tools can be called in parallel, and tool_choice supports auto and none.
For machine-readable results, use JSON mode or provide a strict JSON Schema. This is useful for extraction, routing, evaluation, and agent workflows that require a stable response contract.
Pricing and data use
Muse Spark 1.2 uses pay-as-you-go token pricing, with a separate discounted rate for cached input and an additional per-query rate when built-in web search runs. See the live Muse Spark 1.2 model page and pricing page for current rates.
Meta states that prompts and completions submitted to this Standard checkpoint are not used to train its models. EmpirioLabs does not retain prompt or message content in usage logs.
Start building
Try Muse Spark 1.2 in the Playground, read the API documentation, or compare it with other reasoning models in the model catalog.
