Open Model Lab

Runs

Future run records will connect model variants, eval suites, configs, metrics, caveats, and report links.

Status

No public runs yet. This page defines the planned run registry shape and will stay empty until real runs, configs, datasets, and reports exist.

Future run fields

run id
date
model
model variant
eval suite
dataset version
config hash
score
cost
latency
failure mode distribution
known caveats
report link

Schema example

This is a schema/example, not a real run.

{
  "run_id": "schema_example_not_a_real_run",
  "date": "YYYY-MM-DD",
  "model": "model identifier",
  "model_variant": "base | sft | dpo | agent",
  "eval_suite": "eval_tasks_v1",
  "dataset_version": "version or null",
  "config_hash": "hash or null",
  "score": null,
  "cost": null,
  "latency": null,
  "failure_mode_distribution": {},
  "known_caveats": [],
  "report_link": null
}