GitLab

block on write which only returns when the channel is closed. this is
contrary to the previous arrangement where the handler may return but
the stream hasn't finished writing. it can lead to the client receiving
unexpected responses (since the request has been handled) or worst case
a nil-pointer dereference as the stream tries to flush a nil writer

continue conversation

fix typo

feed responses back into the llm

Timings

fix pull race

fix eof error in generate

error checking new model

offline fixes

call llama.cpp directly from go

writing logs to `./ollama/logs`