What's Changed
- Update dependency huggingface-hub to ~=1.21.0 by @renovate[bot] in #2801
- Add Pi coding agent to ramalama sandbox by @bmahabirbu in #2812
- Add ramalama.conf tempdir setting for TMPDIR selection by @rhatdan in #2781
- Fix import order in bin/ramalama to use local package by @bmahabirbu in #2824
- Add HF_ENDPOINT environment variable support to specify Hugging Face base URL by @olliewalsh in #2826
- fix: pin transformers to workaround mlx-lm issue by @olliewalsh in #2833
- Update dependency huggingface-hub to ~=1.22.0 by @renovate[bot] in #2829
- mask Authorization header in debug log output by @olliewalsh in #2804
- Update CI updates by @renovate[bot] in #2828
- Respect bind host from cli args in health check by @olliewalsh in #2834
- Add ramalama models command to list server-loaded models by @rhatdan in #2830
- feat: add global --engine-args for podman/docker (discussion #2703) by @rhatdan in #2728
- Update quay.io/ascend/cann Docker tag to v9.0.1 by @renovate[bot] in #2840
- Update docusaurus monorepo to v3.10.2 by @renovate[bot] in #2838
- Remove duplicate image args in quantize cli by @olliewalsh in #2844
- Bump llama.cpp to b9986 by @olliewalsh in #2842
- Add url parameter to sandbox commands by @mkristian in #2811
- Update dependency huggingface-hub to ~=1.23.0 by @renovate[bot] in #2841
- update RAG deps and sandbox images to their latest versions by @mikebonnet in #2845
- Use a Hugging Face mirror for CI by @olliewalsh in #2827
- Add speculative decoding CLI args by @olliewalsh in #2848
- Auto-pull draft model in run/serve handlers by @olliewalsh in #2849
- -tools: update library versions and fix torch installation by @mikebonnet in #2850
- handle empty HF_ENDPOINT by @mikebonnet in #2851
- Update CI updates by @renovate[bot] in #2853
- Update CI updates (major) by @renovate[bot] in #2855
- Update dependency huggingface-hub to ~=1.24.0 by @renovate[bot] in #2854
- Add router mode support to ramalama sandbox by @bmahabirbu in #2822
- Add ramalama colours to docs ui by @jessicachitas in #2856
- sandbox: cleanup e2e tests and Markdown by @mikebonnet in #2858
- infer first model from given server: by @mkristian in #2857
- fix(cuda): tell users to install nvidia-container-toolkit when CDI is… by @rhatdan in #2862
- Update actions/checkout action to v7.0.1 - autoclosed by @renovate[bot] in #2864
- Update astral-sh/setup-uv action to v9 by @renovate[bot] in #2866
- Update dependency typescript to v7 by @renovate[bot] in #2869
- Fix docsite build by escaping urls in sandbox manpages by @olliewalsh in #2870
- Update react monorepo to v19.2.8 by @renovate[bot] in #2865
- refactor: consolidate bind host formatting into host_utils by @rhatdan in #2835
- add hosts entry to fix RAG and sandbox under Docker by @mikebonnet in #2872
- only set OpenVINO env vars if the "openvino" backend is selected by @mikebonnet in #2874
- docs: auto-sync version in man pages and track generated markdown by @olliewalsh in #2876
- fix(perplexity): serveral small bugs by @mkristian in #2879
- Update CI updates (major) by @renovate[bot] in #2883
- Update pypa/gh-action-pypi-publish action to v1.14.2 by @renovate[bot] in #2881
- Update step-security/harden-runner action to v2.20.1 by @renovate[bot] in #2886
- Update dependency huggingface-hub to ~=1.27.0 by @renovate[bot] in #2882
- Update step-security/harden-runner action to v2.21.0 by @renovate[bot] in #2891
- Update dependency wheel to ~=0.48.0 by @renovate[bot] in #2890
- gguf_inspect: support NVFP4 format by @ueno in #2894
- Update astral-sh/setup-uv action to v10 by @renovate[bot] in #2892
- add ffmpeg as a runtime dependency for all container images by @olliewalsh in #2885
- SECURITY.md update to follow CRA template by @dominikkawka in #2846
- Bump llama.cpp to v0.1.2 by @olliewalsh in #2896
- ci: work around hanging Azure apt mirror by @olliewalsh in #2900
- update RAG deps and sandbox images to their latest versions by @olliewalsh in #2899
- Fix RAG proxy streaming content-type by @pradhnyesh01 in #2902
- openvino: bump OpenVINO to 2026.2.1 to match llama.cpp v0.1.2 by @olliewalsh in #2904
- Bump version to v0.24.0 by @olliewalsh in #2907
New Contributors
- @jessicachitas made their first contribution in #2856
- @pradhnyesh01 made their first contribution in #2902
Full Changelog: v0.23.0...v0.24.0