---
title: "List checkpoints"
method: GET
path: "/v1/rl/checkpoints"
tags: ["RL"]
---

# List checkpoints

`GET /v1/rl/checkpoints`

## Request

- Base URL: `https://api.together.ai`
- URL: `https://api.together.ai/v1/rl/checkpoints`
- Auth: HTTP bearer

Lists training and inference checkpoints owned by the caller, newest first. Filter by type, session, or base model, for example to recover a checkpoint ID for resume.

## Query parameters

- `session_id` string: Only return checkpoints produced by this training session
- `base_model` string: Only return checkpoints trained from this base model. Match is exact.
- `limit` integer: Maximum number of checkpoints to return (1-100)
- `after` string: Cursor for pagination (ID of the last checkpoint from the previous page)
- `type` 'CHECKPOINT_TYPE_TRAINING' | 'CHECKPOINT_TYPE_INFERENCE': Only return checkpoints of this type. CHECKPOINT_TYPE_TRAINING is the full training state (weights and optimizer state), for resuming a training session with its optimizer state; CHECKPOINT_TYPE_INFERENCE is a model ready for serving or download, also added to your models. When set, it must be CHECKPOINT_TYPE_TRAINING or CHECKPOINT_TYPE_INFERENCE; when omitted, checkpoints of both types are returned.

## Response `200`

List of checkpoints

- RLCheckpointsListResponse: A page of checkpoints
  - `data` RLCheckpoint[], required: Checkpoints in this page
    - `id` string, required: Unique identifier for the checkpoint
    - `session_id` string, required: Training session that produced the checkpoint
    - `base_model` string, required: Base model the checkpoint was trained from
    - `type` 'CHECKPOINT_TYPE_TRAINING' | 'CHECKPOINT_TYPE_INFERENCE', required: Whether a checkpoint is saved for training resume or inference download.
    - `step` union, required: Training step at time of save
      - string
      - integer
    - `lora_rank` integer: LoRA rank of the session that produced this checkpoint. Absent for full-weight sessions and for checkpoints saved before this field was recorded.
    - `created_at` string, date-time, required: Timestamp when the checkpoint was created
    - `inference_registration` RLInferenceCheckpointRegistration: Where an inference checkpoint's weights are stored in the Together model registry. At least one of `model` and `adapter` is set.
      - `model` RLModelRegistryArtifact: A specific revision of a model in the Together model registry
        - `id` string, required: ID of the model in the Together model registry, as used by the models API
        - `revision_id` string, required: Revision of the model that holds this checkpoint
      - `adapter` RLModelRegistryArtifact: A specific revision of a model in the Together model registry
        - `id` string, required: ID of the model in the Together model registry, as used by the models API
        - `revision_id` string, required: Revision of the model that holds this checkpoint
    - `training_registration` RLModelRegistryArtifact: A specific revision of a model in the Together model registry
      - `id` string, required: ID of the model in the Together model registry, as used by the models API
      - `revision_id` string, required: Revision of the model that holds this checkpoint
  - `meta` RLListMeta, required: Pagination metadata for list responses
    - `limit` integer: Maximum number of items returned per page
    - `has_more` boolean: Whether more items exist beyond this page
    - `next_cursor` string: Cursor to use as the 'after' parameter for the next page. Empty when has_more is false.

## Other responses

- `default`: An unexpected error response.

## Changes

- **2026-10-22** `bb5257b3173a`: 1 info
  - endpoint added

[Operation history](https://skmtc.dev/together/apis/together-apis/changes/v1/rl/checkpoints/get.md)

---

[API](https://skmtc.dev/together/apis/together-apis.md) - [All operations](https://skmtc.dev/together/apis/together-apis/llms.txt) - [OpenAPI document](https://skmtc.dev/together/apis/together-apis/revisions/93438ddac711?raw)