Can I use these models for commercial projects?

Yes, but you must comply with the specific license for each model (e.g., Llama Community License). Review the Acceptable Use Policy carefully.

What's the difference between Muse Spark, Muse Glimmer, and Llama?

Muse Spark is a hosted, agentic coding model on the API. Muse Glimmer is an open model for local deployment. Llama is Meta's foundation model family, also available via the API.

Is the API compatible with existing OpenAI SDKs?

Yes, the Meta Model API is designed to be compatible with OpenAI SDKs and libraries for easy integration.

How does prompt caching work?

The API caches prompts to reduce latency and cost on subsequent identical requests. It's a built-in feature of the Responses and Chat Completions API protocols.

Directory/Coding/AI Development Platforms/Meta Model API

Meta Model API

Meta Model API is a direct, self-serve platform providing access to Meta's suite of AI models, including the Muse Spark series for coding and agentic tasks, Muse Image for generation, Muse Voice Transcribe for speech-to-text, and SAM 3.1 for image/video segmentation. It offers capabilities like tool calling, search grounding, and multimodal perception. The API is built for developers needing frontier-level models with OpenAI SDK compatibility for easy integration into existing workflows and coding agents.

Coding

Visit website

Added on

Sep 22, 2026

Muse Spark 1.3Muse GlimmerSAM 3.1Image generationVoice transcriptionVideo understandingLocal deploymentFine-tuning

Product information

Everything worth knowing about Meta Model API

The complete picture — from everyday features to the technical detail builders and IT folks dig for.

Core features

What it actually does

Multi-agent coding in your terminal

Muse Code provides a multi-agent coding environment directly in your terminal, built for agentic workflows with Muse Spark.

Direct API access to Muse models

Get self-serve, affordable access to the full suite of Muse models (Spark, Glimmer, Image, Voice Transcribe, SAM) via the Meta Model API.

Multimodal perception

Muse Spark can perceive and reason over video, images, and documents, executing visual reasoning in a real environment.

Competitive coding performance

Models are tuned for long-horizon coding workflows, delivering cleaner output and fewer unnecessary turns, performing competitively with frontier models.

Technical capabilities

Tool calling and search grounding

Capabilities for integrating tool calls, searching tools, and grounding responses with live web information.

Support for multiple API protocols

API compatible with Responses API, Chat Completions API, and Messages API protocols.

Must watch videos

Introducing SAM Audio: The First Unified Multimodal Model for Audio Separation | AI at Meta

1:09

Tutorial1:09

Introducing SAM Audio: The First Unified Multimodal Model for Audio Separation | AI at Meta

🔉 Introducing SAM Audio, the first unified model that isolates any sound from complex audio mixtures using text, visual, or span prompts. SAM Audio represents a new era in audio separation technology, outperforming previous models across a wide range of benchmarks and tasks. We’re sharing SAM Audio with the community, along with a perception encoder model, benchmarks and research papers, to empower others to explore new forms of expression and build applications that were previously out of reach. 🔗 Learn more: https://go.meta.me/a5e8ef -- Subscribe: https://www.youtube.com/aiatmeta?sub_confirmation=1 Learn more about our work: https://ai.meta.com Follow us on Twitter: https://twitter.com/aiatmeta Follow us on Facebook: https://www.facebook.com/aiatmeta Connect with us on LinkedIn: https://www.linkedin.com/showcase/aiatmeta/ Meta focuses on bringing the world together by advancing AI, powering meaningful and safe experiences, and conducting open research.

Introducing SAM 3D: a New Standard for 3D Object & Human Reconstruction from a Single Image

0:50

Tutorial0:50

Introducing SAM 3D: a New Standard for 3D Object & Human Reconstruction from a Single Image

Introducing SAM 3D, the newest addition to the SAM collection, bringing common sense 3D understanding of everyday images. SAM 3D includes two models: 🛋️ SAM 3D Objects for object and scene reconstruction 🧑‍🤝‍🧑 SAM 3D Body for human pose and shape estimation Both models achieve state-of-the-art performance transforming static 2D images into vivid, accurate reconstructions. 🔗 Learn more: https://go.meta.me/85bb63 -- Subscribe: https://www.youtube.com/aiatmeta?sub_confirmation=1 Learn more about our work: https://ai.meta.com Follow us on Twitter: https://twitter.com/aiatmeta Follow us on Facebook: https://www.facebook.com/aiatmeta Connect with us on LinkedIn: https://www.linkedin.com/showcase/aiatmeta/ Meta focuses on bringing the world together by advancing AI, powering meaningful and safe experiences, and conducting open research.

Introducing Meta Segment Anything Model 3 (SAM 3): Unified Detection, Segmentation & Tracking

1:10

Tutorial1:10

Introducing Meta Segment Anything Model 3 (SAM 3): Unified Detection, Segmentation & Tracking

Meet SAM 3, a unified model that enables detection, segmentation, and tracking of objects across images and videos. SAM 3 introduces some of our most highly requested features like text and exemplar prompts to segment all objects of a target category. Learnings from SAM 3 will help power new features in Instagram Edits and Vibes, bringing advanced segmentation capabilities directly to creators. We’re sharing SAM 3 under the SAM License so others can use it to build their own experiences 🔗 Learn more: https://go.meta.me/f987cd -- Subscribe: https://www.youtube.com/aiatmeta?sub_confirmation=1 Learn more about our work: https://ai.meta.com Follow us on Twitter: https://twitter.com/aiatmeta Follow us on Facebook: https://www.facebook.com/aiatmeta Connect with us on LinkedIn: https://www.linkedin.com/showcase/aiatmeta/ Meta focuses on bringing the world together by advancing AI, powering meaningful and safe experiences, and conducting open research.

Who uses it

Real use cases, no hype

Agentic Coding & Development

Build autonomous coding agents, multi-agent orchestration, and AI pair programming using Muse Spark and Muse Code.

Multi-Modal Application Building

Create apps that process and reason over text, images, video, and voice in a single workflow.

Computer Vision & Media Analysis

Implement object detection, segmentation, and tracking in images and video with SAM 3.1.

Speech-to-Text & Transcription

Build transcription services with speaker diarization and real-time processing using Muse Voice Transcribe.

AI-Powered Image Generation

Generate and refine images via text prompts using the agentic Muse Image model.

On-Device / Edge AI Deployment

Deploy quantized models like Muse Glimmer locally for low-latency, always-on agent applications.

Structured Data Extraction & Reasoning

Use the API for complex reasoning, tool calling, and generating JSON-structured outputs from unstructured data.

What's great

Deep integration with Meta's own AI ecosystem (Llama, SAM, Muse).
Multi-modal capabilities out-of-the-box (text, image, video, voice, segmentation).
Generous free tier for initial development and testing.

Technical strengths

Unified API for multiple specialized models (Muse Spark, Image, Voice).
Native support for agentic workflows, tool calling, and structured output.

Where it falls short

Documentation can feel fragmented across multiple model-specific sections.
Commercial use requires careful review of varying license terms per model.
Complex pricing and rate limit structure across different capabilities.

Technical limitations

SDKs primarily OpenAI-compatible; native SDKs may lag in feature parity.
Some advanced models (e.g., Llama 3.1 70B) may be deprecated or replaced without direct download paths.

For technical folks

The deep-dive specs

FAQ

Includes technical Q&A

Sign up at dev.meta.ai, get an API key, and use the OpenAI-compatible endpoint. The quickstart guide gets you a working request in minutes.

Traffic Insights

Monthly Visits

—

October 2026

Growth Rate

vs last month

Dominance

in Coding

Top Country

—

Top Countries Breakdown

—

Growth Trend

Stable

Traffic has remained stable.

Pricing overview

Freemium Pricing

Meta Model API Pricing

Meta Model API uses a freemium pricing structure. Real-time rates, custom enterprise quotas, and plan details are managed directly on their official platform.

Check Pricing on Meta Model API

Reviews