List training-run checkpoints

GET

https://api.fastino.ai

training

jobs

{job_id}

checkpoints

curl https://api.fastino.ai/v1/training/jobs/YOUR_JOB_ID/checkpoints \
  -H "X-API-Key: $FASTINO_API_KEY"

{
  "success": true,
  "run_id": "YOUR_SESSION_ID:train:0",
  "checkpoints": [
    {
      "checkpoint_id": "sampler_weights/step-100",
      "checkpoint_type": "sampler",
      "time": "2026-10-09T18:40:02Z",
      "tinker_path": "tinker://YOUR_SESSION_ID:train:0/sampler_weights/step-100",
      "size_bytes": 167772160,
      "public": false,
      "user_metadata": null,
      "expires_at": null
    },
    {
      "checkpoint_id": "weights/step-100",
      "checkpoint_type": "training",
      "time": "2026-10-09T18:40:00Z",
      "tinker_path": "tinker://YOUR_SESSION_ID:train:0/weights/step-100",
      "size_bytes": 503316480,
      "public": false,
      "user_metadata": null,
      "expires_at": null
    }
  ],
  "count": 2
}

GET /v1/training/jobs/{job_id}/checkpoints Returns the saved checkpoints of a training run, newest first. Checkpoints are created by the SDK’s save calls: training-state checkpoints (weights/<name>) can resume training, and sampler checkpoints (sampler_weights/<name>) hold the adapter weights used for sampling.

Request

string

required

Your Fastino API key.

string

required

Training-run ID.

curl https://api.fastino.ai/v1/training/jobs/YOUR_JOB_ID/checkpoints \
  -H "X-API-Key: $FASTINO_API_KEY"

Response

boolean

Whether the request succeeded.

string

Training-run ID.

object[]

Saved checkpoints, newest first.

Show checkpoint

string

Checkpoint ID, either weights/<name> or sampler_weights/<name>. Use it with Download a checkpoint.

string

training for training-state checkpoints or sampler for sampler weights.

string

ISO 8601 timestamp when the checkpoint was saved.

string

The checkpoint’s tinker:// path, which the SDK uses to load it.

integer | null

Checkpoint size in bytes, when known.

boolean

Whether the checkpoint is published.

object | null

String key-value metadata attached to the checkpoint.

string | null

ISO 8601 expiry time, or null when the checkpoint is kept until deleted.

integer

Number of checkpoints returned.

{
  "success": true,
  "run_id": "YOUR_SESSION_ID:train:0",
  "checkpoints": [
    {
      "checkpoint_id": "sampler_weights/step-100",
      "checkpoint_type": "sampler",
      "time": "2026-10-09T18:40:02Z",
      "tinker_path": "tinker://YOUR_SESSION_ID:train:0/sampler_weights/step-100",
      "size_bytes": 167772160,
      "public": false,
      "user_metadata": null,
      "expires_at": null
    },
    {
      "checkpoint_id": "weights/step-100",
      "checkpoint_type": "training",
      "time": "2026-10-09T18:40:00Z",
      "tinker_path": "tinker://YOUR_SESSION_ID:train:0/weights/step-100",
      "size_bytes": 503316480,
      "public": false,
      "user_metadata": null,
      "expires_at": null
    }
  ],
  "count": 2
}

An unknown run, or a run you cannot see, returns 404.

⌘I

curl https://api.fastino.ai/v1/training/jobs/YOUR_JOB_ID/checkpoints \
  -H "X-API-Key: $FASTINO_API_KEY"

{
  "success": true,
  "run_id": "YOUR_SESSION_ID:train:0",
  "checkpoints": [
    {
      "checkpoint_id": "sampler_weights/step-100",
      "checkpoint_type": "sampler",
      "time": "2026-10-09T18:40:02Z",
      "tinker_path": "tinker://YOUR_SESSION_ID:train:0/sampler_weights/step-100",
      "size_bytes": 167772160,
      "public": false,
      "user_metadata": null,
      "expires_at": null
    },
    {
      "checkpoint_id": "weights/step-100",
      "checkpoint_type": "training",
      "time": "2026-10-09T18:40:00Z",
      "tinker_path": "tinker://YOUR_SESSION_ID:train:0/weights/step-100",
      "size_bytes": 503316480,
      "public": false,
      "user_metadata": null,
      "expires_at": null
    }
  ],
  "count": 2
}

​Request

​Response

Request

Response