Skip to main content
GET
/
models
List models
curl --request GET \
  --url https://api.overshoot.ai/v1/models
{
  "object": "list",
  "data": [
    {
      "id": "Qwen/Qwen3.5-9B",
      "object": "model",
      "created": 123,
      "owned_by": "overshoot",
      "status": "ready"
    }
  ]
}

Documentation Index

Fetch the complete documentation index at: https://docs.overshoot.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
What’s in the response

  • Each entry’s id is what you pass as model on /chat/completions. Use it verbatim — IDs are case-sensitive.
  • status is the only mutable field worth checking:
    • ready — at least one healthy replica is serving. Use it.
    • loading — replicas are starting; may flip to ready in seconds, may not. Don’t depend on it.
    • Models with no healthy replica are omitted from the response entirely.
  • A model that’s ready here can still return 503 on /chat/completions if its replica falls over between calls. Always handle 503 with a retry or a fallback to another ready model.
This endpoint is the only one that doesn’t require an API key. List it before every session — don’t hardcode model IDs.

Response

200 - application/json

List of models.

​
object
string
default:list
required

Always list.

​
data
object[]
required

Models that currently have at least one healthy replica or are loading.