---
title: "List models"
method: GET
path: "/models"
---

# List models

`GET /models`

List the live model catalog, including pricing, modalities, and capability metadata.

## Response `200`

Model catalog response

- ModelListResponse
  - `object` string, required — Response container type.
  - `data` ModelRecord[], required — Live model records.
    - `id` string, required — Model ID to pass as model.
    - `object` string, required
    - `created` integer, nullable — Creation timestamp when available.
    - `owned_by` string, nullable
    - `tier` string, nullable — Access tier when applicable.
    - `model_type` string — Broad model family or endpoint type. Video-capable models use video.
    - `pricing_mode` string — Pricing calculation mode. Token-priced models use token; video models use second.
    - `pricing` ModelPricing
      - `input` number — Input-token price for token-priced models.
      - `output` number — Output-token price for token-priced models.
      - `price` number — Per-unit price for non-token pricing modes. Video models use public USD price per generated second.
      - `minimum_request_price_usd` number
      - `minimum_cache_tokens` number
      - `currency` string
      - `unit` string
    - `modalities` ModelModalities
      - `input` string[]
      - `output` string[]
    - `context_window` object, nullable
    - `usage_based_only` boolean
    - `stream` boolean
    - `json_mode` boolean
    - `reasoning` boolean
    - `tools_calling` boolean
    - `capabilities` ModelCapabilities
      - `video_generation` boolean — Whether the model can generate video.
      - `video_async` boolean — Whether video generation is asynchronous.
      - `supported_seconds` integer[] — Allowed seconds values for video generation.
      - `supported_sizes` string[] — Allowed output sizes for video generation.
      - `max_reference_images` integer — Maximum number of uploaded reference images.
      - `max_reference_image_bytes` integer — Maximum bytes per uploaded reference image.
      - `requires_reference_image` boolean — Whether at least one input_reference file is required.

## Other responses

- `429` — Gateway or upstream rate limit.
- `503` — Upstream provider is temporarily unavailable.

---

[API](https://skmtc.dev/llm7/apis/llm7-api.md) · [All operations](https://skmtc.dev/llm7/apis/llm7-api/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/llm7/llm7-api/revisions/2933587838e9/schema)