GitLab

Until we get all the memory calculations correct, this can provide
and escape valve for users to workaround out of memory crashes.

* Added community link for Ollama Copilot

* Update README.md

---------
Co-authored-by: Michael <mchiang0610@users.noreply.github.com>

Add ollama user to video group

Add env var so podman will map cuda GPUs

Omit build date from gzip headers

instead of appending image tags, prepend them - this generally produces better results

On OpenSUSE, ollama needs to be a member of the video group
to access the GPU

Without this env var, podman's GPU logic doesn't map the GPU through

See https://reproducible-builds.org/ for why this is good.

This patch was done while working on reproducible builds for openSUSE.

Bump llama.cpp to b2276

Refine container image build script

Log unexpected server errors checking for update

This should unmask some failure modes that likely
show up in app logs as unmarshal errors

Allow overriding the platform, image name, and tag latest for
standard and rocm images.

* read iogpu.wired_limit_mb on macOS

Fix for https://github.com/ollama/ollama/issues/1826

* improved determination of available vram on macOS

read the recommended maximal vram on macOS via Metal API

* Removed macOS-specific logging

* Remove logging from gpu_darwin.go

* release Core Foundation object

fixes a possible memory leak

specfied -> specified

Remove unused GPT4all
Use nomic-embed-text as embedded model
Fix a deprecation warning (__call__)

remove format private key

this is unnecessary now that x/crypto/ssh.MarshalPrivateKey has been
added

Co-authored-by: bennhuang <bennhuang@tencent.com>

- Enchanted is now supported for desktop on macOS