SYLIN

LM Studio

Runtime · Text & code

Works as a local OpenAI-compatible backend for datasette-llm and Datasette Agent

Evidence

  1. Aug 31
    Commenter uses the Pi agent harness in Docker with Qwen3.8-27B served from LM Studio daily, including PDF editing
  2. Aug 15
    Works with Prime Agent by swapping in the LM Studio endpoint and model name in the same config
  3. Jul 18
    Can run the Gemma 4 QAT build

Alternatives

llama.cppReviewer prefers plain llama.cpp over the Ollama and LM Studio wrappers; lower memory use once context is reduced