Model selection

ColabHive keeps model selection explicit: inspect the task, dataset and available hardware, then choose a compatible training configuration. There is no separate Auto-Mode contract.

Use these resources:

• Choosing a Model — heuristics by task type, dataset size and hardware.
• Model Catalog — model capabilities and runtime requirements.
• Hyperparameter Tuning — per-model parameter reference.

List active training configurations with the SDK:

from colabhive import ColabHive

client = ColabHive(api_key="...", account_id="...")

configs = client.training.model_configs(active_only=True)
for config in configs:
    print(config.model_id, config.display_name, config.framework)

Or query the REST API:

curl "https://api.colabhive.com/api/builder/v1/training/model-configs" \
  -H "X-Account-ID: YOUR_ACCOUNT_ID" \
  -H "X-API-Key: YOUR_API_KEY"

The returned catalog is the source of truth for supported configurations. Validate model licenses, dataset rights and hardware fit before starting a run.