xpost

youtube link

Field.ms

switch to ZeroGPU: transformers pipeline, drop llama-cpp-python

fix segfault: serialize llama.cpp inference with a lock

model update

llmcpppatch

update

first