Skip to content

Instantly share code, notes, and snippets.

@MoserMichael

MoserMichael/using-claude-locally-on-mac.txt

Last active March 10, 2026 04:54
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save MoserMichael/8e6dd03acf910b7c4e8f1f51734a94d1 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save MoserMichael/8e6dd03acf910b7c4e8f1f51734a94d1 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
using-claude-locally-on-mac.txt
# use brew on the mac, this installs ollama that has been compiled with the metal library
# metal is a library that uses the native GPU on the M3 processor
brew install ollama
# check that metal is in use
# if metal is installed correctly, you will see a line like this on the screen:
# "inference compute" ... library=Metal ...
OLLAMA_DEBUG=1 ollama serve
# next thing we want to do: load the claude model
# now stop ollama serve and run it again with the following option.
# environment variable GODEBUG=x509negativeserial=1 tells the http go library to to ignore problems with certificates
# (can occur due to some proxy on the way, when loading claude)
# you can also add OLLAMA_DEBUG=1 if you want to see lots of debugging output: OLLAMA_DEBUG=1 GODEBUG=x509negativeserial=1 ollama serve
GODEBUG=x509negativeserial=1 ollama serve
# in a separate shell window, start pulling the claude model - this one takes quite a lot of time.
# here ollama is talking to the locally running claude server
ollama pull glm-4.7-flash
# after loading the model: run the model now (talks to the claude instance hosted by of $(ollama serve)
# this one asks for a prompt in the terminal, runs reasonably fast on the GPU of an M3 processor!
olama run glm-4.7-flash
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment