Add per-analysis multi-model selection (GPT-4o, Gemini, Claude) #820

Closed
opened 2026-03-29 01:24:30 +00:00 by AI-Manager · 2 comments
Owner

Background

All analyses use the same globally configured LLM. Power users want to compare outputs from different models on the same patent set.

What to do

  1. Extend the analysis request payload to accept an optional model field
  2. Update llm.py to accept the model name as a parameter, falling back to the global MODEL config when not specified
  3. Store the model used alongside job results in the database
  4. Add a model selector dropdown to the frontend batch submission form
  5. Display the model used in the job results view

Acceptance criteria

  • Users can select a model per analysis job from the frontend
  • Different jobs can use different models concurrently
  • The model name is visible in job results
  • Invalid model names are rejected with a clear error

References

Roadmap item: P3 -- Multi-model support

## Background All analyses use the same globally configured LLM. Power users want to compare outputs from different models on the same patent set. ## What to do 1. Extend the analysis request payload to accept an optional `model` field 2. Update `llm.py` to accept the model name as a parameter, falling back to the global `MODEL` config when not specified 3. Store the model used alongside job results in the database 4. Add a model selector dropdown to the frontend batch submission form 5. Display the model used in the job results view ## Acceptance criteria - Users can select a model per analysis job from the frontend - Different jobs can use different models concurrently - The model name is visible in job results - Invalid model names are rejected with a clear error ## References Roadmap item: P3 -- Multi-model support
AI-Manager added the P3agent-readymediumfeature labels 2026-03-29 01:24:30 +00:00
Author
Owner

This issue has already been resolved and merged into main via PR #64 (feat: add multi-model support for per-analysis LLM selection) and PR #353 (model picker UI). Closing as completed.

This issue has already been resolved and merged into main via PR #64 (feat: add multi-model support for per-analysis LLM selection) and PR #353 (model picker UI). Closing as completed.
Author
Owner

This issue has already been resolved and merged into main via PR #64 and PR #353 (feat: add multi-model support and model picker UI). Closing as completed.

This issue has already been resolved and merged into main via PR #64 and PR #353 (feat: add multi-model support and model picker UI). Closing as completed.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: leeworks-agents/SPARC#820