What's Changed
Breaking Changes 🛠
- 🔥 add LLaVA support and GPT vision API, Multiple requests for llama.cpp, return JSON types by @mudler in #1254
- refactor: rename llama-stable to llama-ggml by @mudler in #1287
Bug fixes 🐛
- fix: respect OpenAI spec for response format by @mudler in #1289
- fix: handle grpc and llama-cpp with REBUILD=true by @mudler in #1328
- fix: propagate CMAKE_ARGS when building grpc by @mudler in #1334
Exciting New Features 🎉
- feat(certificates): add support for custom CA certificates by @vitorstone in #880
- feat(conda): conda environments by @mudler in #1144
- refactor: move backends into the backends directory by @mudler in #1279
- feat: allow to run parallel requests by @mudler in #1290
- feat(transformers): add embeddings with Automodel by @mudler in #1308
- ci(core): add -core images without python deps by @mudler in #1309
- feat: initial watchdog implementation by @mudler in #1341
👒 Dependencies
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1231
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1236
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1285
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1288
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1291
Other Changes
- Update .gitignore for backend/llama.cpp by @dave-gray101 in #1235
- llama index example by @sfxworks in #1237
- chianlit example by @sfxworks in #1238
- Fixes the bug 1196 by @diego-minguzzi in #1232
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1242
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1256
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1265
- deps(go-piper): update to 2023.11.6-3 by @M0Rf30 in #1257
- feat(llama.cpp): support lora with scale and yarn by @mudler in #1277
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1272
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1280
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1282
- feat: queue up requests if not running parallel requests by @mudler in #1296
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1297
- fix(api/config): allow YAML config with .yml by @Papawy in #1299
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1300
- llava.yaml (yaml format standardization) by @lunamidori5 in #1303
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1304
- ⬆️ Update mudler/go-piper by @localai-bot in #1305
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1306
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1310
- fix: ExLlama Backend Context Size & Rope Scaling by @ok2sh in #1311
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1313
- docs: Initial import from localai-website by @mudler in #1312
- fix: move python header comments below shebang in some backends by @B4ckslash in #1321
- Feat: OSX Local Codesigning by @dave-gray101 in #1319
- docs: Add llava, update hot topics by @mudler in #1322
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1323
- docs: Update Features->Embeddings page to reflect backend restructuring by @B4ckslash in #1325
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1330
- fix: rename transformers.py to avoid circular import by @mudler in #1337
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1340
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1345
- feat(petals): add backend by @mudler in #1350
- fix: go-piper add libucd at linking time by @M0Rf30 in #1357
- docs: Add docker instructions, add community projects section in README by @mudler in #1359
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1351
- docs: Update getting started and GPU section by @mudler in #1362
- ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1363
- ci: limit concurrent jobs by @mudler in #1364
- fix/docs: Python backend dependencies by @B4ckslash in #1360
- ci: split into reusable workflows by @mudler in #1366
- fix: OSX Build Fix Part 1: Metal by @dave-gray101 in #1365
- docs: add fine-tuning example by @mudler in #1374
- docs: site/how-to clean up by @lunamidori5 in #1342
New Contributors
- @vitorstone made their first contribution in #880
- @sfxworks made their first contribution in #1237
- @diego-minguzzi made their first contribution in #1232
- @M0Rf30 made their first contribution in #1257
- @Papawy made their first contribution in #1299
- @ok2sh made their first contribution in #1311
- @B4ckslash made their first contribution in #1321
Full Changelog: v1.40.0...v2.0.0_beta