API reference
Models
GET /v1/models — list the models available to the bearer key.
GET/v1/models
| Base URL | https://litellm.tensorloop.tech |
| Auth | Authorization: Bearer YOUR_KEY |
Response
OpenAI-shaped:
{
"object": "list",
"data": [
{ "id": "gpt-4o-mini", "object": "model", "owned_by": "openai" }
]
}The list is filtered by your plan — free returns only gpt-4o-mini; pro returns the full set the upstream proxy exposes.
Available models
| Model | Plan | Context | Modalities |
|---|---|---|---|
gpt-4o-mini | Free Pro | 128k | text, vision |
gpt-4o | Pro | 128k | text, vision |
claude-3-5-sonnet | Pro | 200k | text, vision |
llama-3.1-70b-instruct | Pro | 128k | text |
mixtral-8x7b-instruct | Pro | 32k | text |
The upstream catalog evolves — hit /v1/models for the canonical list at any time. Pricing is set per upstream model and is metered through your key budget.
Per-key vs per-plan
This endpoint returns the models your plan allows, not the subset your specific key is scoped to. A pro user with a narrowly-scoped key (e.g. only claude-3-5-sonnet) still sees the full pro catalog here.
To check what a specific key can call, look at its Models column in the dashboard — see Dashboard walkthrough.
Errors
401— missing or revoked key.502— upstream proxy unavailable.