models

name	models
description	Browse, compare, evaluate, and manage trained models. View leaderboard, trends, lineage, and artifacts. Deploy models to Ollama. Use when asked about models, model comparison, which model is best, model evaluation, deploying a model, or downloading a model.

Models

Browse and manage the model registry.

List Models

scripts/api.sh GET /models

Leaderboard

Ranked models by evaluation score:

scripts/api.sh GET /models/leaderboard

Trends

Model performance over time:

scripts/api.sh GET /models/trends

Compare Models

scripts/api.sh POST /models/compare '{"model_ids": ["model-a", "model-b"]}'

Get Model Details

scripts/api.sh GET /models/{model_id}

Evaluate a Model

Run evaluation benchmarks:

scripts/api.sh POST /models/{model_id}/evaluate '{}'

Deploy to Ollama

Make a model available locally via Ollama:

scripts/api.sh POST /models/{model_id}/deploy-ollama '{}'

Delete a Model

Destructive — confirm with user first.

scripts/api.sh DELETE /models/{model_id}

Model Artifacts

View training artifacts (configs, checkpoints):

scripts/api.sh GET /models/{model_id}/artifacts

Example Interactions

User: "Which model is best?" → Call GET /models/leaderboard. Report top 3 with scores.

User: "Compare model-a and model-b" → Call POST /models/compare. Present side-by-side metrics.

User: "Deploy my latest model to Ollama" → Call GET /models to find latest. Call POST /models/{id}/deploy-ollama.