What's Changed
- Bump to v0.7.5 by @rhatdan in #1233
- ad-hoc llama clustering option by @afazekas in #1238
- Fix: huggingface logout doesn't use token by @melodyliu1986 in #1237
- Fix Typo in Clustering Placeholder Comment by @afazekas in #1240
- Allow building older versions of cuda by @rhatdan in #1235
- Always use --jinja with run too by @afazekas in #1241
- Fix for using client/server version of "ramalama run" by @ericcurtin in #1242
- support AI Models environment variables doc2rag/rag_framework by @rhatdan in #1243
- docs: update CONTRIBUTING.md section on bots by @nathan-weinberg in #1245
- Deleting default values for issues by @ericcurtin in #1249
- preload docling models by @bmahabirbu in #1246
- Add gawk by @ericcurtin in #1252
- Allow jinja argument by @ericcurtin in #1250
- Verify OCI image format in rag command by @rhatdan in #1253
- Move model and rag to use shared Engine implementation by @rhatdan in #1236
- Deleting default values for bug reports by @ericcurtin in #1256
- Set model store as default by @engelmi in #1189
- Change default testing to use --pull=missing by @rhatdan in #1254
- We need fixes around CPU support etc. by @ericcurtin in #1265
- Automatically pick cuda docker container by @edmcman in #1266
- Add minor changes to enhance pylint mark by @sarroutbi in #1276
- Use --quiet for ramalama version by @rhatdan in #1272
- Replace project.license with an SPDX license expression. by @jwieleRH in #1271
- Enhance pylint mark in ramalama/cli.py by @sarroutbi in #1277
- Initial draft model support by @afazekas in #1255
- Fix formatting as suggested by isort so that "make validate" passes. by @jwieleRH in #1270
- Fix up description of draft model by @rhatdan in #1279
- Add information on configuring the libkrun machine provider by @rhatdan in #1273
- Remove hardcodeing to ecurtin $HOME by @ericcurtin in #1280
- Revert "Replace project.license with an SPDX license expression." by @ericcurtin in #1281
- Remove model tag from file name on migration by @engelmi in #1283
- Use RamaLama instead of Ramalama for consistency by @sarroutbi in #1286
- Fix minor issues by @sarroutbi in #1285
- Add gemma3 shortnames by @ericcurtin in #1288
- Workaround for CUDA image not pointing to libcuda.so.1 in ld.so.conf by @olivergs in #1287
- Change CLI behaviour by @ericcurtin in #1293
- Exit ramalama if user types exit by @ericcurtin in #1291
- Fix to check correct directory for shortnames file by @ericcurtin in #1290
- Use current ramalama directory rather them main from repo by @rhatdan in #1282
- Add shortname information to ramalama info by @rhatdan in #1295
- Make --no-cache optional for make build by @rhatdan in #1294
- Update registry.access.redhat.com/ubi9/ubi Docker tag to v9.5-1745848351 by @renovate in #1300
- Bump to v0.8.0 by @rhatdan in #1284
New Contributors
- @sarroutbi made their first contribution in #1276
- @jwieleRH made their first contribution in #1271
- @olivergs made their first contribution in #1287
Full Changelog: v0.7.5...v0.8.0