Meta's updated frontier reasoning model with a 1,048,576-token context, image, video, audio, and PDF understanding, web search, and tool calling.
Meta AIText Generation1M contextReleased Aug 5, 2026Proprietary EndpointNewMeta's updated frontier reasoning model with a 1,048,576-token context, image, video, audio, and PDF understanding, web search, and tool calling.
Reasoning is always on and cannot be disabled. Reasoning stays internal, and reasoning tokens are billed as output tokens and count against max_tokens. reasoning_effort accepts minimal through xhigh. Text, image, MP4 video, MP3/WAV audio, and PDF inputs are supported; up to 50 images can be attached. PDFs use text from the first 100 pages and page images from the first 50 pages. Built-in web search adds $0.00825 per executed search query and reports the count in usage.tool_usage. OpenAI Responses requests also support custom tools and deferred tool search. tool_choice supports auto and none. Prompts and completions sent to this Standard checkpoint are not used by Meta to train its models.
Also known as Muse Spark, Meta AI Muse Spark 1.2
muse-spark-1-2/v1/chat/completionsPOST/v1/responsesPOST/v1/messagesPOST/v1beta/models/muse-spark-1-2:generateContentmuse-spark-1.2meta/muse-spark-1-2meta/muse-spark-1.2Live pay-as-you-go rates from the EmpirioLabs catalog. You are billed only for what you use, with no monthly minimum.
Muse Spark 1.2 serves the OpenAI-compatible Chat Completions API. Point any OpenAI SDK at https://api.empiriolabs.ai/v1 with your EmpirioLabs API key and use the model id muse-spark-1-2. Get an API key from the EmpirioLabs dashboard.
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",
"messages": [
{"role": "user", "content": "Write a haiku about the ocean."}
]
}'from openai import OpenAI
client = OpenAI(
base_url="https://api.empiriolabs.ai/v1",
api_key="YOUR_EMPIRIOLABS_API_KEY",
)
response = client.chat.completions.create(
model="muse-spark-1-2",
messages=[{"role": "user", "content": "Write a haiku about the ocean."}],
)
print(response.choices[0].message.content)Request parameters supported by the Muse Spark 1.2 API on EmpirioLabs. Defaults apply when a field is omitted.
| Parameter | Type | Default | Range / values | Description |
|---|---|---|---|---|
| max_tokens | integer | 16384 | 1 to 131072 | Maximum number of output tokens to generate. Reasoning tokens count against this budget. |
| temperature | number | 1 | 0 to 2 | Controls randomness. Lower values make responses more deterministic. |
| top_p | number | 1 | 0.01 to 1 | Nucleus sampling cutoff. |
| presence_penalty | number | 0 | -2 to 2 | Penalizes tokens that already appeared, encouraging new topics. |
| frequency_penalty | number | 0 | -2 to 2 | Penalizes frequent tokens, reducing repetition. |
| seed | integer | - | - | Random seed for more reproducible sampling. |
| reasoning_effort | enum | medium | minimal, low, medium, high, xhigh | Reasoning is always on; this sets how much effort the model spends before answering. Higher effort uses more reasoning tokens. Reasoning text is not returned in... |
| tool_web_search | boolean | false | - | Enable built-in web search with cited sources. Adds $0.00825 per executed search query; a single request can run more than one search. |
| tools | array | [] | - | OpenAI-compatible function and custom tool definitions. On /v1/responses, tool_search and defer_loading can discover deferred tools. |
| tool_choice | object | - | - | OpenAI-compatible tool choice control. This model supports auto and none. |
| parallel_tool_calls | boolean | true | - | Allow the model to request multiple function tools in one response. |
| response_format | enum | - | - | Return structured JSON output. JSON mode returns any valid JSON object; JSON Schema mode enforces an exact schema. |
Reasoning is always on and cannot be disabled. Reasoning stays internal, and reasoning tokens are billed as output tokens and count against max_tokens. reasoning_effort accepts minimal through xhigh. Text, image, MP4 video, MP3/WAV audio, and PDF inputs are supported; up to 50 images can be attached. PDFs use text from the first 100 pages and page images from the first 50 pages. Built-in web search adds $0.00825 per executed search query and reports the count in usage.tool_usage. OpenAI Responses requests also support custom tools and deferred tool search. tool_choice supports auto and none. Prompts and completions sent to this Standard checkpoint are not used by Meta to train its models.
When this model invokes tools inside a single request, the response carries a normalized usage.tool_usage map alongside the token counts. The example below shows the shape. Exact field names, units, and which tools appear can vary by tool:
"usage": {
"prompt_tokens": 123,
"completion_tokens": 456,
"cost_usd": 0.0042,
"tool_usage": {"web_search": 3}
}The tool counts are already factored into cost_usd and are surfaced so you can audit per-tool billing. The field is omitted when no billed tool was invoked.
On EmpirioLabs, Muse Spark 1.2 is billed pay as you go: Input $1.25 per 1M prompt tokens; Output $4.25 per 1M generated tokens; Implicit cache read $1.00 per 1M cached input tokens. The live rate card on this page always matches what the API charges.
Muse Spark 1.2 supports a 1M-token context window with up to 131,072 output tokens per response.
Yes. Muse Spark 1.2 serves the OpenAI-compatible Chat Completions API, so existing OpenAI SDKs work by pointing base_url at https://api.empiriolabs.ai/v1 and setting the model id to muse-spark-1-2.
Yes. The EmpirioLabs playground runs Muse Spark 1.2 in the browser with the same parameters the API exposes, so you can test prompts before writing code.
Create an EmpirioLabs account, then generate a key under API Keys in the dashboard. Billing is pay-as-you-go credits, so you only pay for the requests you make.
Check out our pricing or reach out if you want your own model deployed on our stack.
