Tag: 로컬LLM
All the articles with the tag "로컬LLM".
-
Ollama Local LLM Routing #2 — I was talking to a local model using OpenAI API format
I opened up code I had only set up but never read last time. Tracing it from the pydantic models in types.py to the abstract class in backends, I found out that requests sent to the local ollama were in OpenAI API format.
-
Ollama Local LLM Routing #1 — One big model would do, but I split it into four small ones
I decided to try routing prompts across several open-source LLMs on my MacBook first. Both my home computer and MacBook fell short on specs, so I had to pick different models, and I finished installing ollama.