llm 9
- Benchmark subagent models the way you run them
- Doctrine prose degrades; deterministic hooks don't
- Running Qwen 3.6 35B MoE on an RTX 3060 12GB via -ncmoe
- 1.5× Faster Agentic Coding with MTP on Qwen 3.6 27B
- issuer PR Dashboard: Web UI on Top of SQLite + LLM Reviews
- Extending Issuer to Pull Requests: Same Pattern, Different Data
- Running Issuer at Scale: Closing 42 Issues Without Losing Control
- issuer: A Local SQLite-Backed GitHub Issue Manager Powered by LLMs
- Running NVIDIA Nemotron 30B with Vision on a 24 GB GPU