Puntos de control interactivos
Guardar pesos de entrenamiento
Guarda con un nombre los pesos LoRA y el estado del optimizador de un trabajo interactivo. Devuelve un future_id, el ID de weights y su ruta tinker://.
POST
POST /v1/training/jobs/{job_id}/save-weights
Saves the job’s LoRA weights and optimizer state as weights/<name>. You can
load it back into the same job,
or start a new job from it. To sample
or download weights, save them with
POST /v1/training/jobs/{job_id}/save-weights-for-sampler
instead.
The save runs in order with the job’s other calls.
Request
string
requerido
Your Fastino API key.
string
requerido
Interactive job ID.
string
requerido
1–128 characters from
A-Z, a-z, 0-9, ., _, and -.integer | null
Delete the checkpoint automatically after this many seconds, from
3600
(1 hour) to 315360000 (10 years). Omit it to keep the checkpoint until you
delete it. Live checkpoints count toward storage usage.boolean
predeterminado:"false"
Replace a state checkpoint with the same name.
object
String key-value metadata stored with the checkpoint.
Response
string
The
save future.string
weights/<name>.string
Full checkpoint path. Pass it as
from_checkpoint when you create a job.