LI Model Router: ask once, compare answers from open-source AI models

Auto Route

Hello! Just ask. A router runs a few suitable models, and you pick the answer to continue with.

Manual Chat

Hello! Select a model in the panel and start typing.

Thinking

Reasoning Effort

Temperature

0.20

Max Tokens

Prompt

Sent to candidate models

Evaluation

Used by the judge to score outputs

Candidate Models Fan-out runs all selected models in parallel

Evaluate candidate models in parallel: rubric-based scoring, independent evals, full trace logging.

Select all models

Chat History

Every Manual Chat turn is saved here: text, uploaded images, and generated images.

Trace Explorer

All runs append full outputs, token usage, served_by backends, and independent judge verdicts to JSONL.

Timestamp	User	Query	Goal	Candidates	Achieved Goal	Judge Model	Status
Loading trace logs…