Skip to main content
GET
GET /v1/training/jobs/{job_id}/futures/{future_id} Every interactive-job call that does work returns a future_id. Read the future until its status is completed, failed, or cancelled. Use wait=true to wait at most 25 seconds per request. Continue polling a nonterminal future. Use wait=true to wait at most 25 seconds per request. Continue polling a nonterminal future. Use wait=true to wait at most 25 seconds per request. Continue polling a nonterminal future. Use wait=true to wait at most 25 seconds per request. Continue polling a nonterminal future. Use wait=true to wait at most 25 seconds per request. Continue polling a nonterminal future. Results are kept for one hour.

Request

string
requerido
Your Fastino API key.
string
requerido
Interactive job ID.
string
requerido
Future ID returned when the work was queued.
boolean
predeterminado:"false"
Wait up to 25 seconds for an unfinished future before answering. Repeat the request until it finishes.

Response

string
The future.
string
The job.
string
create, load, forward, forward_backward, optim_step, save, or sample.
string
pending, running, completed, failed, or cancelled.
string | null
While unfinished: active, or paused_capacity while GPU capacity warms up.
any
Once completed:
  • forward and forward_backward: loss_fn_outputs (one entry of tensors per datum, such as logprobs) and metrics.
  • sample: sequences, each with tokens, logprobs, and stop_reason, plus prompt_logprobs when requested.
  • save: the checkpoint path.
  • Other types: an acknowledgement.
object | null
When failed or cancelled: message, and category (user for a problem with the request, server for a Fastino problem).