FAfastino-ai
training

Save an interactive training job's training state

Changed on

Save the LoRA weights and optimizer state under a name, to load later or start a new job from (Tinker's save_state). Read the result at GET /v1/training/jobs/{job_id}/futures/{future_id}.

post/v1/training/jobs/{job_id}/save-weights

Request

Path parameters

job_idstring required

Headers

Idempotency-Keystring nullable

Client-generated key (a UUID is ideal; it must be URL-safe to look it up) that makes the call safe to send again after a lost response. A repeat with the same key and body from the same caller returns the first call's answer (the same job and future) without running or billing it again, or a 409 while the first call still runs; the same key with another route, job, body or caller is a 422. Look a key up with GET /v1/training/operations/{idempotency_key}.

Request body

namestring required

1-128 characters from [A-Za-z0-9._-].

overwriteboolean

Replace a state of the same name.

ttl_secondsinteger nullable

1 hour to 10 years; none keeps it until deleted.

user_metadataobject nullable

Response

Successful Response

checkpoint_idstring required

weights/<name> or sampler_weights/<name>.

future_idstring required
job_idstring required
pathstring required

The checkpoint's tinker:// path.

typestring required

The call: create, load, forward, forward_backward, optim_step, save, sample.

Changes