---
meta:
title: Models
description: Available models on Meta Model API and guidance on choosing the right model for your workload.
keywords: models, Muse Spark, contributor tier, pricing tier, model selection, model lineup, capabilities, context window
cms:
alias: /model-api/docs/models
target: aidmc
---
# Models
Meta Model API runs Meta's latest models for agentic and coding work — multi-step tool loops, software engineering assistants, and long-context reasoning. This page shows what's available today and what each model can do.
## Available models {#available-models}
| Model ID | Tier | Input modalities | Output modalities | Context window |
| :---- | :---- | :---- | :---- | :---- |
| `muse-spark-1.1` | [Standard](/docs/pricing-rate-limits#standard-tier) | Text, image, video, PDF | Text | 1,048,576 tokens |
| `muse-spark-1.2` | [Standard](/docs/pricing-rate-limits#standard-tier) | Text, image, video, PDF | Text | 1,048,576 tokens |
| `muse-spark-1.2-contributor` | [Contributor](/docs/pricing-rate-limits#contributor-tier) | Text, image, video, PDF | Text | 1,048,576 tokens |
All three IDs serve the Muse Spark family and share modalities and a 1,048,576-token context window. They differ by checkpoint and **tier**:
- **`muse-spark-1.1`** — the earlier checkpoint, on the **[Standard](/docs/pricing-rate-limits#standard-tier)** tier.
- **`muse-spark-1.2`** — an updated checkpoint with slightly higher performance, also on the **[Standard](/docs/pricing-rate-limits#standard-tier)** tier.
- **`muse-spark-1.2-contributor`** — the `muse-spark-1.2` checkpoint on the discounted **[Contributor](/docs/pricing-rate-limits#contributor-tier)** tier, where your prompts and completions may be used to train future Meta models.
`muse-spark-1.2` is the default model in the code examples throughout these docs.
Muse Spark is multimodal: it takes text, image, video, audio, and PDF as input and generates text. Use it for [chat completion](/docs/protocols/chat-completions), [image understanding](/docs/image-understanding), [video and audio understanding](/docs/video-understanding), [tool calling](/docs/tool-calling), [structured output](/docs/structured-output), and [search grounding](/docs/search-grounding).
### Tiers {#tiers}
Tier is a model attribute: it sets the price you pay and whether your data may be used to train future Meta models.
- **[Standard](/docs/pricing-rate-limits#standard-tier)** (`muse-spark-1.1`, `muse-spark-1.2`) — standard pricing; your prompts and completions are not used to train Meta models.
- **[Contributor](/docs/pricing-rate-limits#contributor-tier)** (`muse-spark-1.2-contributor`) — heavily discounted pricing in exchange for permission to use your prompts and completions to train future Meta models.
The [contributor tier](/docs/pricing-rate-limits#contributor-tier) lowers the barrier to entry: it gives you room to prototype, test integrations, and scale experiments without the usual cost overhead, in return for permission to train on your data.
See [Pricing and rate limits](/docs/pricing-rate-limits) for per-tier pricing and model availability.
## List models via the API {#list-models}
Query the catalog programmatically when you need to check what's enabled for your team:
```python title="Python (OpenAI SDK)"
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.meta.ai/v1",
api_key=os.environ["MODEL_API_KEY"],
)
response = client.models.list()
print(response.model_dump_json(indent=2))
```
```python title="Python (requests)"
import json
import os
import requests
response = requests.get(
"https://api.meta.ai/v1/models",
headers={"Authorization": f"Bearer {os.environ['MODEL_API_KEY']}"},
)
response.raise_for_status()
print(json.dumps(response.json(), indent=2))
```
```shell title="curl"
curl -X GET "https://api.meta.ai/v1/models" \
-H "Authorization: Bearer $MODEL_API_KEY"
```
The response returns models sorted newest-first by creation time, with the model ID as a stable tiebreaker. Each object includes a `created` field — the Unix timestamp in seconds for when the model was added to the registry. `GET /v1/models/{model}` returns the same `created` value for a single model.
## Next steps
- [Get started](/docs/quickstart#first-call): make your first call with Muse Spark
- [Pricing and rate limits](/docs/pricing-rate-limits): see rates and retry guidance
- [Chat completion](/docs/protocols/chat-completions): start generating with the core conversational endpoint
Models
Meta Model API runs Meta's latest models for agentic and coding work — multi-step tool loops, software engineering assistants, and long-context reasoning. This page shows what's available today and what each model can do.Model ID | Tier | Input modalities | Output modalities | Context window |
|---|
muse-spark-1.1
| | Text, image, video, PDF | Text | 1,048,576 tokens |
muse-spark-1.2
| | Text, image, video, PDF | Text | 1,048,576 tokens |
muse-spark-1.2-contributor
| | Text, image, video, PDF | Text | 1,048,576 tokens |
- •
muse-spark-1.1 — the earlier checkpoint, on the Standard tier. - •
muse-spark-1.2 — an updated checkpoint with slightly higher performance, also on the Standard tier. - •
muse-spark-1.2-contributor — the muse-spark-1.2 checkpoint on the discounted Contributor tier, where your prompts and completions may be used to train future Meta models.
muse-spark-1.2 is the default model in the code examples throughout these docs.Muse Spark is multimodal: it takes text, image, video, audio, and PDF as input and generates text. Use it for chat completion, image understanding, video and audio understanding, tool calling, structured output, and search grounding.Tier is a model attribute: it sets the price you pay and whether your data may be used to train future Meta models.- •Standard (
muse-spark-1.1, muse-spark-1.2) — standard pricing; your prompts and completions are not used to train Meta models. - •Contributor (
muse-spark-1.2-contributor) — heavily discounted pricing in exchange for permission to use your prompts and completions to train future Meta models.
created field — the Unix timestamp in seconds for when the model was added to the registry. GET /v1/models/{model} returns the same created value for a single model.