Skip to content

Releases: mudler/LocalAI

v2.4.0

04 Jan 16:30
@mudler mudler
v2.4.0
bcf0244
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.4.0

What's Changed

Bug fixes 🐛

  • fix(entrypoint): cd to backend dir before start by @mudler in #1530
  • fix(diffusers): add omegaconf dependency by @mudler in #1540

Exciting New Features 🎉

👒 Dependencies

Other Changes

Full Changelog: v2.3.1...v2.4.0

Contributors

mudler and localai-bot
Assets 8
Loading

v2.3.1

30 Dec 14:38
@mudler mudler
v2.3.1
a95bb05
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.3.1

What's Changed

Bug fixes 🐛

  • Fix: Set proper Homebrew install location for x86 Macs by @cnatale in #1510
  • fix(download): correctly check for not found error by @mudler in #1514

👒 Dependencies

Other Changes

New Contributors

Full Changelog: v2.3.0...v2.3.1

Contributors

cnatale, mudler, and 2 other contributors
Assets 8
Loading

v2.3.0

29 Dec 19:01
@mudler mudler
v2.3.0
9723c3c
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.3.0

What's Changed

Bug fixes 🐛

Exciting New Features 🎉

👒 Dependencies

Other Changes

New Contributors

Full Changelog: v2.2.0...v2.3.0

Contributors

M0Rf30, sozercan, and 5 other contributors
Assets 8
Loading

v2.2.0

21 Dec 16:48
@mudler mudler
v2.2.0
9ae47d3
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.2.0

What's Changed

This release brings updates to the backends and includes a fix for recompilation of LocalAI with go-rwkv (#1473). To note, it also tries to reduce the image size by allowing some backends (transformers-based) to share the same environment.

With this release inline templates and models as URLs are supported, for example:

name: mixtral

parameters:
  model: https://huggingface.co/TheBloke/Mixtral-8x7B-v0.1-GGUF/resolve/main/mixtral-8x7b-v0.1.Q2_K.gguf
# or huggingface://TheBloke/Mixtral-8x7B-v0.1-GGUF/mixtral-8x7b-v0.1.Q2_K.gguf@main
template:
  completion: |
            Complete the following: {{.Input}}

Exciting New Features 🎉

  • docs: automatically track latest versions by @mudler in #1451
  • feat: inline templates and accept URLs in models by @mudler in #1452
  • feat(conda): share envs with transformer-based backends by @mudler in #1465

👒 Dependencies

Other Changes

Full Changelog: v2.1.0...v2.2.0

Contributors

mudler, lunamidori5, and localai-bot
Assets 8
Loading

v2.1.0

16 Dec 21:04
@mudler mudler
v2.1.0
3d83128
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.1.0

What's Changed

Breaking Changes 🛠

  • feat(alias): alias llama to llama-cpp, update docs by @mudler in #1448

Bug fixes 🐛

  • fix(piper): pin petals, phonemize and espeak by @mudler in #1393
  • update(llama.cpp): update server, correctly propagate LLAMA_VERSION by @mudler in #1440

Exciting New Features 🎉

👒 Dependencies

Other Changes

New Contributors

Full Changelog: v2.0.0...v2.1.0

Contributors

sozercan, mudler, and 4 other contributors
Assets 8
Loading

v2.0.0

04 Dec 13:22
@mudler mudler
v2.0.0
238fec2
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.0.0

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
  • fix(vall-e-x): correctly install reqs in environment by @mudler in #1377

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
  • feat: update whisper_cpp with CUBLAS, HIPBLAS, METAL, OPENBLAS, CLBLAST support by @wuxxin in #1302

👒 Dependencies

Other Changes

New Contributors

Full Changelog: v1.40.0...v2.0.0

Contributors

M0Rf30, mudler, and 10 other contributors
Assets 8
Loading
0 Join discussion

v2.0.0_beta

01 Dec 19:16
@mudler mudler
v2.0.0_beta
d6073ac
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
v2.0.0_beta Pre-release
Pre-release

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

Other Changes

New Contributors

Full Changelog: v1.40.0...v2.0.0_beta

Contributors

M0Rf30, mudler, and 9 other contributors
Assets 8
Loading

v1.40.0

30 Oct 18:34
@mudler mudler
v1.40.0
6ef7ea2
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
v1.40.0

This release is a preparation before v2 - the efforts now will be to refactor, polish and add new backends. Follow up on: #1126

Hot topics

This release now brings the llama-cpp backend which is a c++ backend tied to llama.cpp. It follows more closely and tracks recent versions of llama.cpp. It is not feature compatible with the current llama backend but plans are to sunset the current llama backend in favor of this one. This one will be probably be the latest release containing the older llama backend written in go and c++. The major improvement with this change is that there are less layers that could be expose to potential bugs - and as well it ease out maintenance as well.

Support for ROCm/HIPBLAS

This release bring support for AMD thanks to @65a . See more details in #1100

More CLI commands

Thanks to @jespino now the local-ai binary has more subcommands allowing to manage the gallery or try out directly inferencing, check it out!

What's Changed

Bug fixes 🐛

  • fix(openai): Populate ID and Created fields in OpenAI compatible responses by @jespino in #1164
  • Fix backend/cpp/llama CMakeList.txt on OSX by @dave-gray101 in #1212

Exciting New Features 🎉

👒 Dependencies

  • fix(deps): update module github.com/onsi/gomega to v1.28.0 by @renovate in #1113
  • ⬆️ Update go-skynet/go-llama.cpp by @localai-bot in #1106
  • fix(deps): update github.com/tmc/langchaingo digest to e16b777 by @renovate in #1101
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to 79f9587 by @renovate in #1085
  • fix(deps): update module github.com/shirou/gopsutil/v3 to v3.23.9 by @renovate in #1120
  • fix(deps): update module github.com/sashabaranov/go-openai to v1.15.4 by @renovate in #1122
  • fix(deps): update module github.com/rs/zerolog to v1.31.0 by @renovate in #1102
  • ⬆️ Update go-skynet/go-llama.cpp by @localai-bot in #1130
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to 6018c9d by @renovate in #1129
  • ⬆️ Update go-skynet/go-llama.cpp by @localai-bot in #1136
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to 1676dcd by @renovate in #1135
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to 56c0d28 by @renovate in #1140
  • fix(deps): update module github.com/onsi/ginkgo/v2 to v2.13.0 by @renovate in #1152
  • fix(deps): update module google.golang.org/grpc to v1.58.3 by @renovate in #1160
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to aeba71e by @renovate in #1155
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to 10f9b49 by @renovate in #1158
  • fix(deps): update module github.com/sashabaranov/go-openai to v1.16.0 by @renovate in #1159
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to 22de3c5 by @renovate in #1172
  • fix(deps): update github.com/tmc/langchaingo digest to a02d4fd by @renovate in #1175
  • fix(deps): update module github.com/gofiber/fiber/v2 to v2.50.0 by @renovate in #1177
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to 9a19c74 by @renovate in #1179
  • fix(deps): update github.com/tmc/langchaingo digest to c636b3d by @renovate in #1188
  • fix(deps): update module google.golang.org/grpc to v1.59.0 by @renovate in #1189
  • chore(deps): update actions/checkout action to v4 by @renovate in #1006
  • feat(llama.cpp): update by @mudler in #1200
  • ⬆️ Update go-skynet/go-llama.cpp by @localai-bot in #1156
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to c25dc51 by @renovate in #1191
  • ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1204
  • fix(deps): update module github.com/onsi/gomega to v1.28.1 by @renovate in #1205

Other Changes

  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to 6711bdd by @renovate in #1079
  • ci: cleanup worker by @mudler in #1166
  • docs(examples): Add mistral example by @mudler in #1214
  • feat(llama.cpp): Bump llama.cpp, adapt grpc server by @mudler in #1211
  • cleanup: drop bloomz and ggllm as now supported by llama.cpp by @mudler in #1217
  • ci: use self-hosted to build container images by @mudler in #1206
  • ci: run only cublas on selfhosted by @mudler in #1224
  • ⬆️ Update ggerganov/llama.cpp by @localai-bot in #1207

New Contributors

Full Changelog: v1.30.0...v1.40.0

Contributors

jespino, renovate, and 6 other contributors
Assets 8
Loading

v1.30.0

25 Sep 16:49
@mudler mudler
v1.30.0
274ace2
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
v1.30.0

What's Changed

This is an exciting LocalAI release! Besides bug-fixes and enhancements this release brings the new backend to a whole new level by extending support to vllm and vall-e-x for audio generation!

Bug fixes 🐛

  • fix(llama): resolve lora adapters correctly from the model file by @mudler in #964
  • fix(diffusers): correctly check alpha by @mudler in #967
  • fix(utf8): prevent multi-byte utf8 characters from being mangled by @samm81 in #981
  • fix(vall-e-x): make audiopath relative to models by @mudler in #1012
  • fix(vall-e-x): copy vall-e-x next to the local-ai binary in the container image by @mudler in #1082

Exciting New Features 🎉

  • feat(diffusers): Add lora by @mudler in #965
  • feat: expose all the jobs via /models/jobs endpoint by @Jirubizu in #983
  • feat: Model Gallery Endpoint Refactor / Mutable Galleries Endpoints by @dave-gray101 in #991
  • Allow to manually set the seed for the SD pipeline by @maxjcohen in #998
  • feat: add vall-e-x by @mudler in #1007
  • feat(vllm): Initial vllm backend implementation by @mudler in #1026
  • feat(speculative-sampling): allow to specify a draft model in the model config by @mudler in #1052
  • fix(vllm): set default top_p with vllm by @mudler in #1078
  • feat(python-grpc): allow to set max workers with PYTHON_GRPC_MAX_WORKERS by @mudler in #1081
  • feat(vllm): Allow to set quantization by @mudler in #1094

👒 Dependencies

  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to 9072315 by @renovate in #963
  • ⬆️ Update go-skynet/go-llama.cpp by @ci-robbot in #968
  • fix(deps): update module github.com/gofiber/fiber/v2 to v2.49.0 by @renovate in #966
  • fix(deps): update github.com/tmc/langchaingo digest to c85d396 by @renovate in #962
  • fix(deps): update module github.com/valyala/fasthttp to v1.49.0 by @renovate in #971
  • fix(deps): update go-llama.cpp by @mudler in #980
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to bf3f946 by @renovate in #979
  • fix(deps): update module github.com/sashabaranov/go-openai to v1.15.1 by @renovate in #984
  • fix(deps): update module github.com/shirou/gopsutil/v3 to v3.23.8 by @renovate in #989
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to b6e38d6 by @renovate in #988
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to c5622a8 by @renovate in #992
  • ⬆️ Update go-skynet/go-llama.cpp by @ci-robbot in #996
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to d8c8547 by @renovate in #997
  • ⬆️ Update go-skynet/go-llama.cpp by @ci-robbot in #1002
  • fix(deps): update module github.com/gofiber/fiber/v2 to v2.49.1 by @renovate in #1001
  • chore(deps): remove unused package by @appleboy in #1003
  • ⬆️ Update go-skynet/go-llama.cpp by @ci-robbot in #1005
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to 05dc4b6 by @renovate in #1004
  • feat(llama.cpp): update by @mudler in #1024
  • fix(deps): update module github.com/sashabaranov/go-openai to v1.15.2 by @renovate in #1022
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to 4145bd5 by @renovate in #1025
  • ⬆️ Update go-skynet/go-llama.cpp by @ci-robbot in #1027
  • fix(deps): update github.com/tmc/langchaingo digest to 9c8845b by @renovate in #1029
  • ⬆️ Update go-skynet/go-llama.cpp by @ci-robbot in #1043
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to f0735ef by @renovate in #1034
  • fix(deps): update module github.com/sashabaranov/go-openai to v1.15.3 by @renovate in #1035
  • chore(deps): update docker/build-push-action action to v5 by @renovate in #1039
  • chore(deps): update docker/login-action action to v3 by @renovate in #1040
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to cc8a123 by @renovate in #1041
  • ⬆️ Update go-skynet/go-llama.cpp by @ci-robbot in #1048
  • chore(deps): update docker/metadata-action action to v5 by @renovate in #1045
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to cf4eb53 by @renovate in #1047
  • ⬆️ Update go-skynet/go-llama.cpp by @ci-robbot in #1062
  • fix(deps): update module google.golang.org/grpc to v1.58.1 by @renovate in #1020
  • fix(deps): update module github.com/gofiber/fiber/v2 to v2.49.2 by @renovate in #1049
  • fix(deps): update module github.com/valyala/fasthttp to v1.50.0 by @renovate in #1060
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to b471eb7 by @renovate in #1050
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to e86c637 by @renovate in #1059
  • ⬆️ Update go-skynet/go-llama.cpp by @ci-robbot in #1084
  • fix(llama.cpp): update, run go mod tidy by @mudler in #1088
  • fix(deps): update module google.golang.org/grpc to v1.58.2 by @renovate in #1090
  • fix(deps): update github.com/tmc/langchaingo digest to 2c309cf by @renovate in #1097

Other Changes

  • [query_data example] max_chunk_overlap in PromptHelper must be in 0..1 range by @quoing in #1000
  • feat(diffusers): don't set seed in params and respect device by @mudler in #1010
  • Update README to reflect changes in Continue's config file by @rdeaton in #1014
  • 1038 - Streamlit bot with LocalAI by @majoshi1 in #1072
  • Cleaned up chatbot-ui READMEs by @jamesbraza in #1075
  • Requested Changes from GPT4ALL to Luna-AI-Llama2 by @lunamidori5 in #1092
  • Add the CONTRIBUTING.md by @Aisuko in #1098

New Contributors

Full Changelog: v1.25.0...v2.0.0

Contributors

appleboy, majoshi1, and 12 other contributors
Assets 8
Loading
0 Join discussion

v1.25.0

26 Aug 14:43
@mudler mudler
v1.25.0
9e5fb29
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
v1.25.0

What's Changed

Breaking Changes 🛠

  • feat: bump llama.cpp, add gguf support by @mudler in #943

Bug fixes 🐛

  • fix: disable usage by default (still experimental) by @mudler in #929
  • fix: drop racy code, refactor and group API schema by @mudler in #931

Exciting New Features 🎉

  • feat(Makefile): allow to restrict backend builds by @mudler in #890
  • feat(diffusers): various enhancements by @mudler in #895
  • feat: make initializer accept gRPC delay times by @mudler in #900
  • feat(diffusers): add DPMSolverMultistepScheduler++, DPMSolverMultistepSchedulerSDE++, guidance_scale by @mudler in #903
  • feat(diffusers): overcome prompt limit by @mudler in #904
  • feat(diffusers): add img2img and clip_skip, support more kernels schedulers by @mudler in #906
  • Usage Features by @dave-gray101 in #863
  • feat(diffusers): be consistent with pipelines, support also depthimg2img by @mudler in #926
  • feat: add --single-active-backend to allow only one backend active at the time by @mudler in #925
  • feat: add llama-stable backend by @mudler in #932
  • feat: allow to customize rwkv tokenizer by @dave-gray101 in #937
  • feat: backend monitor shutdown endpoint, process based by @dave-gray101 in #938
  • feat: Allow to load lora adapters for llama.cpp by @mudler in #955

👒 Dependencies

  • ⬆️ Update nomic-ai/gpt4all by @ci-robbot in #878
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to 4e55940 by @renovate in #893
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to 18f25c2 by @renovate in #894
  • ⬆️ Update nomic-ai/gpt4all by @ci-robbot in #899
  • fix(deps): update github.com/tmc/langchaingo digest to eb0cbd3 by @renovate in #902
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to f03869d by @renovate in #901
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to a630935 by @renovate in #898
  • fix(deps): update module github.com/shirou/gopsutil/v3 to v3.23.7 - autoclosed by @renovate in #924
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to 36f7fb5 by @renovate in #908
  • fix(deps): update github.com/tmc/langchaingo digest to fef0821 by @renovate in #922
  • ⬆️ Update nomic-ai/gpt4all by @ci-robbot in #911
  • fix(deps): update module github.com/google/uuid to v1.3.1 by @renovate in #936
  • fix(deps): update module github.com/onsi/ginkgo/v2 to v2.12.0 by @renovate in #949
  • fix(deps): update github.com/nomic-ai/gpt4all/gpt4all-bindings/golang digest to 27a8b02 by @renovate in #947
  • ⬆️ Update nomic-ai/gpt4all by @ci-robbot in #953
  • ⬆️ Update go-skynet/go-llama.cpp by @ci-robbot in #954
  • fix(deps): update github.com/go-skynet/go-llama.cpp digest to bf63302 by @renovate in #930
  • fix(deps): update github.com/tmc/langchaingo digest to 1e2a401 by @renovate in #948
  • fix(deps): update module github.com/otiai10/openaigo to v1.6.0 by @renovate in #960

Other Changes

New Contributors

Full Changelog: v1.24.1...v1.25.0

Contributors

renovate, mudler, and 3 other contributors
Assets 8
Loading