BeefAPIChoose a model
Match model IDs to text, image, and video endpoints and check current availability and USD prices.
Model IDs identify both the model and the API shape you should use. A text request, an image request, and a video task are different calls.
| Task | Example model ID | Endpoint | Guide |
|---|---|---|---|
| Chat and text | gpt-5.6-sol | /v1/chat/completions or /v1/responses | API reference |
| Anthropic Messages | claude-sonnet-4-6 | /v1/messages | API reference |
| Generate an image | gpt-image-2 | /v1/images/generations | Images |
| Edit an image | gpt-image-2 | /v1/images/edits | Images |
| Generate a video | grok-imagine-video-1.5 | /v1/videos | Videos |
These are example IDs, not a fixed list of everything available. Use the exact spelling returned for your key. Do not send an image model to Chat Completions or assume all models support every endpoint.
List models for your key
Base URL: https://global.beefapi.com/v1
curl --fail-with-body https://global.beefapi.com/v1/models \
-H "Authorization: Bearer $BEEFAPI_KEY"The response contains object: "list" and a data array. Each entry's id is a model ID. Without authentication, this endpoint returns the public catalog rather than your key's permissions.
A catalog entry is not a guarantee of immediate capacity. Your key's restrictions, current availability, and configured price also affect whether a request succeeds. If an ID is listed but returns 503, check Status and Errors; do not repeatedly change a correctly spelled ID.
Check prices and capabilities
Pricing and pricing.usd.json use the same global USD pricebook. Text prices are per million tokens; image prices are per image; video prices are per second at the selected resolution.
Model names alone do not establish context length, tool support, accepted image sizes, or video limits. Use the relevant guide above and the response to your actual request. For a capability not documented here, ask support with the model ID and intended endpoint before depending on it.