# Meta Model API (slim)

> Meta Model API is Meta's pay-as-you-go API for its hosted Muse models, including Muse Wonder and Muse Spark, through OpenAI-style Responses and Chat Completions endpoints and an Anthropic-style Messages endpoint, plus image, speech and segmentation routes.

- Full: https://www.anchorterminal.com/tools/meta-model-api.md (~7,350 tokens) · this version ~1,780 tokens · JSON https://www.anchorterminal.com/tools/meta-model-api.json · canonical https://www.anchorterminal.com/tools/meta-model-api
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-12

**D · 52.2/100 · rank #757 of 993 · #16 in Model APIs & inference · not agent-ready · confidence medium**


Assessment: Meta Model API serves Muse Wonder, in preview since 12 October 2026, and Muse Spark with 1M-token context, constrained JSON Schema output, automatic prompt caching and per-token prices published without a login. Standard-mode content is not used for training. There's no free allowance, status history, OpenAPI file, deprecation policy or way to force a tool call.

## Facts

- Kind: Model API · vendor: Meta · category: Model APIs & inference · legal entity: Meta Platforms, Inc. · provenance 63/100
- Endpoint: `https://api.meta.ai/v1` (HTTP)
- Auth: API key · pricing: Pay per use · x402: no · licence: Closed service under the Meta Model API Terms of Service (last updated 9 October 2026), with Meta Platforms, Inc., or Meta Platforms Ireland Limited in the European Region, as the contracting party.
- Probe metrics: not measured yet (probes haven't run)
- Surface graded: The Meta Model API at api.meta.ai/v1, read from its docs, pricing, help centre, terms and policies on 12 October 2026, with Muse Wonder as the model in question. No authenticated call was made. An unauthenticated `GET /v1/models` answered 401 and `GET /v1/status` answered 200 with `service_status` operational.
- Muse Wonder: Model ID muse-wonder-1.0, in preview. Text, image, video, audio and PDF in, text out, 1,048,576-token context, reasoning effort from minimal to max. Standard mode only on the API. Muse Wonder Full is not on the API.
- Muse Spark: muse-spark-1.3 (recommended for new work), muse-spark-1.2 and muse-spark-1.1, with contributor variants muse-spark-1.3-contributor and muse-spark-1.2-contributor. Same inputs and 1,048,576-token context.
- Other models: muse-image-1.0 for image generation and editing at $0.01 an image, muse-voice-transcribe-1.0 for speech to text at $0.18 an hour of audio, and sam-3.1 for segmentation at $2.50 per 1,000 images or $0.20 per 1,000 video frames.
- Endpoints: Responses (with background mode), Chat Completions, Messages, token counting, Files, Models, image generation and edits, file transcription, a realtime transcription WebSocket at wss://api.meta.ai/v1/asr/realtime, and an unauthenticated Status endpoint.
- Rate limits: Per team and per model. muse-wonder-1.0 and muse-spark-1.3 at 3,000 RPM and 4,000,000 TPM, the contributor variants at 150 RPM and 3,000,000 TPM, and new background responses at 600 a minute per team.
- Pricing modes: Standard mode, where content is not used for training, and Contributor mode, at lower prices in exchange for training use, chosen by model ID. Contributor mode isn't available for Muse Wonder on the API.
- Data handling: Content and usage data are retained as needed to provide the service, for the law, for Meta's rights under the terms, for flagged policy violations and for safety and security review, with no period stated. Zero data retention for qualified accounts through sales, with flagged inputs and outputs kept up to 2 years.
- Availability: The geographic use policy (last updated 12 October 2026) lists 21 restricted territories, among them China including Hong Kong and Macau, Russia and Pakistan, and excludes India from Muse Image.
- Prices: Muse Wonder input $4 per 1M tokens; Muse Wonder cached input $0.20 per 1M tokens; Muse Wonder output $20 per 1M tokens; Muse Spark input $1.25 per 1M tokens; Muse Spark cached input $0.15 per 1M tokens; Muse Spark output $4.25 per 1M tokens; Muse Spark input, Contributor mode $0.10 per 1M tokens; Muse Spark output, Contributor mode $0.20 per 1M tokens; Web search grounding $2.50 per 1,000 requests; Muse Image generated image $0.01 per image

- Scores: Reliability 40, Performance pending, Schema & documentation 60, Agent ergonomics 75, Security & auth 64, Payments & pricing 20, Task success pending, Maintenance & community 46, Transparency & trust 52 · total over the 7 assessed categories
- Why: Reliability, No status page with component history. · Schema & documentation, No OpenAPI or similar machine-readable contract found. · Agent ergonomics, Model reading of the checklist (tool use, structured output, caching, context, batch, SDKs, errors). · Security & auth, Model reading. · Payments & pricing, No machine payment protocol (0). · Maintenance & community, Model reading. · Transparency & trust, Closed service with clear terms, under California law with AAA arbitration for Meta Platforms, Inc.
- Sources: 24, open questions: 7, both in the full twin
- Capabilities: inference.llm, image.generate, image.edit, speech.stt, speech.streaming
- JSON: https://www.anchorterminal.com/api/v1/tools/meta-model-api.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/meta-model-api.svg` or a link to https://www.anchorterminal.com/tools/meta-model-api from a page on meta.ai or one of its subdomains, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Use `muse-wonder-1.0` for long multi-step work and `muse-spark-1.3` for cheaper, high-volume calls. Both take the same requests
2. Keep personal or confidential data away from `-contributor` model IDs. Meta may train on that content
3. Honour Retry-After on 429 and 503, which is 60 seconds for `service_overloaded`, and check `GET https://api.meta.ai/v1/status`, which needs no key
4. Pass the key to the OpenAI SDK explicitly or set `OPENAI_API_KEY`. The SDK doesn't read `MODEL_API_KEY`
5. Leave `tool_choice` at `auto`. The docs say `none` and `required` are not supported

## Connect

```bash
curl -X POST "https://api.meta.ai/v1/responses" \
  -H "Authorization: Bearer $MODEL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "muse-spark-1.3", "input": "What is the capital of France?"}'
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| Cloudflare Workers AI | B | 68.3 | inference.llm, image.generate, speech.stt | https://www.anchorterminal.com/tools/cloudflare-workers-ai.min.md |
| DeepInfra | B | 63 | inference.llm, image.generate, speech.stt | https://www.anchorterminal.com/tools/deepinfra.min.md |
| SiliconFlow | D | 46.7 | inference.llm, image.generate, speech.stt | https://www.anchorterminal.com/tools/siliconflow.min.md |
| Amazon Transcribe | BB | 73.4 | speech.stt, speech.streaming | https://www.anchorterminal.com/tools/amazon-transcribe.min.md |
| Azure AI Speech speech-to-text | BB | 73 | speech.stt, speech.streaming | https://www.anchorterminal.com/tools/azure-speech-to-text.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)