This release adds 4 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+7 more
Summary
AI summaryAuto-lock routes after three corrections make learned routing decisions permanent across sessions.
Full changelog
What's New
Personal Routing Memory System
- Learn from corrections: The router remembers when you use
llm_rerouteto override a decision - Auto-lock routes: After 3+ corrections for the same task type, learned routes become hard overrides in future sessions
- Session memory banner: See your learned routes at session start with confidence levels
- Community sharing: Share your routing profiles via
llm_share_profile()/llm_import_profile(url)
Key Features
- Persistent learned routes stored in
~/.llm-router/learned_routes.json - Session-start hook displays learned patterns with confidence thresholds
- Auto-route hook applies learned overrides before normal classification
- Full test coverage: 11 new tests, 1009 tests passing
Bugs Fixed
- Critical hotfix: Added missing
import pytestin test_hook_health.py
Upgrade
```bash
pip install --upgrade claude-code-llm-router
llm-router install
```
What This Means for You
Your routing decisions are now stateful. When you override a model choice 3 times for the same task type, the router learns and applies it automatically in future sessions. This enables personalized, persistent routing that adapts to your workflow and preferences.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About ypollak2/llm-router
Subscription-aware LLM router for Claude Code. Routes tasks to 20+ providers (OpenAI, Gemini, Groq, Ollama, Codex) based on complexity classification, Claude subscription pressure, and cost. Free tasks stay on Claude subscription; expensive tasks fall back to the cheapest capable model. Includes 30 MCP tools, 6 auto-routing hooks, semantic dedup cache, prompt caching, daily spend cap, and a live web dashboard.
Related context
Related tools
Beta — feedback welcome: [email protected]