What's Changed
- ramalama rm should require at least one argument by @rhatdan in #516
- Fix handling of missing args for login/logout by @rhatdan in #517
- Add --version and fix sort order of commands by @rhatdan in #518
- Update shortnames.conf to alias new granite models by @ericcurtin in #519
- Prettify and improve download bars during model pulls by @swarajpande5 in #521
- Remove
tqdmdependency by @swarajpande5 in #523 - Relocate
FileandHttpClientto their respective modules by @swarajpande5 in #522 - Add missing files to install.sh script by @ericcurtin in #524
- Change pip to pipx in installation instructions by @ericcurtin in #520
- Pass ASAHI, CUDA, HIP, HSA prefixed env vars to container by @abn in #526
- Resolves - Debug option shows traceback when no subcommand is passed by @pghole in #527
- Skip codespell of
venvby adding--skipin the command by @swarajpande5 in #528 - Improved command handling and refactoring by @ericcurtin in #531
- Update README and model.py for Nvidia GPU support by @ericcurtin in #530
- Add UI support documentation by @ericcurtin in #529
- Add llama-run by @ericcurtin in #452
- Update container images to use latest by @ericcurtin in #534
- Update llama.cpp to get context size fix by @ericcurtin in #538
- If runtime is vllm, use image from modh quay.io by @ericcurtin in #533
- Remove huggingface warning by @ericcurtin in #540
- When using podman run add --pull=newer to update container image by @rhatdan in #541
- Build to build_rm by @ericcurtin in #543
- whisper.cpp now installs properly by @rhatdan in #544
- Re-add mkdir -p by @ericcurtin in #545
- This might temporarily fix an issue we are having by @ericcurtin in #547
- Attempt to use versioned container by @ericcurtin in #537
- Fix llama-run to be installed with correct library paths. by @rhatdan in #549
- Add CPU inferencing container build steps to CI by @ericcurtin in #551
- This should save some space by @ericcurtin in #552
- Bump to v0.4.0 by @rhatdan in #550
New Contributors
Full Changelog: v0.3.0...v0.4.0