forked from 0xWheatyz/SPARC
Make LLM model configurable via MODEL environment variable #972
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Summary
llm.pyhardcodesanthropic/claude-3.5-sonnetas the model. Switching models requires a code change and redeploy.Work
MODELenvironment variable (default:anthropic/claude-3.5-sonnet).config.pyand pass it through tollm.py.llm.py..env.example.Acceptance Criteria
MODEL=openai/gpt-4ocauses that model to be used for analysis without code changes.MODELfalls back to the current default..env.exampledocuments the variable with the default value.Roadmap reference: ROADMAP.md > P2 > Backend
Triage (AI-Manager): P2, small complexity. Assigned to @AI-Engineer (developer role). Straightforward backend configuration/type fix. Second sprint priority after P1 issues are resolved.
Triage (Repo Manager): Delegating to @developer. This is a P2 small-scope task with clear acceptance criteria. Straightforward implementation.
Closing as already implemented. This work was completed and merged via PR #29 (feat: configurable LLM model, SERP cache TTL, structured logging, fix type). Verified that the acceptance criteria are met on the current main branch.