New features
- feat: add spark v2 support by @takatost in #885
- feat: server xinference support by @takatost in #927
- feat: add openllm support by @takatost in #928
What's Changed
- feat: support app rename and make app card ui better by @iamjoel in #766
- fix: promptEng menu in wrong place by @iamjoel in #864
- fix: get_num_tokens_from_messages params error by @takatost in #866
- fix multi thread app context by @JohnJyong in #868
- feat: optimize error record in agent by @takatost in #869
- Update README_CN.md by @Panmuse in #867
- Update README.md by @Panmuse in #865
- feat: optimize override app model config convert by @takatost in #874
- feat: optimize model when app create by @takatost in #875
- Feat/customizable file upload config by @MatriQ in #818
- build: improve dockerfile by @MatriQ in #851
- feat: optimize prompt of suggested_questions_after_answer by @takatost in #881
- fix: web dockfile by @zxhlyh in #883
- fix: wenxin error not raise when stream mode by @takatost in #884
- feat: add spark v2 support by @takatost in #885
- feat: support show model display name by @iamjoel in #887
- fix: error raise status code not exist by @takatost in #888
- Feat/pay modal by @zxhlyh in #889
- fix: max token not exist in generate summary when calc rest tokens by @takatost in #891
- feat: claude paid optimize by @takatost in #890
- Fix/embedding chat by @zxhlyh in #899
- fix entrypoint script line endings by @JunIce in #900
- fix: sensitive_word_avoidance npe by @takatost in #902
- fix: react router agent direct output by @takatost in #904
- fix: universal chat when default model invalid by @takatost in #905
- feat: optimize output parse by @takatost in #906
- Feat/segment add tag by @zxhlyh in #907
- Feature/mutil embedding model by @JohnJyong in #908
- feat: optimize service api authorization header invalid error by @takatost in #910
- feat: remove unuse envs by @takatost in #912
- fix: run model serval ex not return by @takatost in #915
- fix: run extra model serval ex not return by @takatost in #916
- fix: azure embedding model credentials include base_model_name is invalid for openai sdk by @takatost in #917
- Fix/price calc by @lingfengchencn in #862
- fix: embedding price config by @takatost in #918
- feat: record price unit in messages by @takatost in #919
- fix: embedding get price definition missing by @takatost in #922
- fix: replicate text generation model validate by @takatost in #923
- feat: optimize generate name length by @takatost in #924
- fix: document paused at format error by @takatost in #925
- Fix Readme.md typo error. by @BenjaminX in #926
- feat: server xinference support by @takatost in #927
- feat: add openllm support by @takatost in #928
- fix: xinference max_tokens alisa error by @takatost in #929
- feat: add xinference embedding model support by @takatost in #930
- fix: remove openllm pypi package because of this package too large by @takatost in #931
- fix: web reader tool missing nodejs by @takatost in #932
- feat: optimize web reader summary in 3.5 by @takatost in #933
- Feat/token support by @JohnJyong in #909
- Feat/add document status tooltip by @zxhlyh in #937
- fix: xinference chat support by @takatost in #939
- fix: login improvement by @MatriQ in #941
- fix: default language improvement in activation page by @MatriQ in #942
- chore: perfect type definition by @jiaowoxiaobala in #940
- fix: add flex-wrap to categories container by @MatriQ in #944
- fix: openllm generate cutoff by @takatost in #945
- fix: provider model not delete when reset key pair by @takatost in #946
- Fix/speech to text button by @zxhlyh in #947
- update document segment by @JohnJyong in #948
- fix: segment error tip & save segment disable when loading by @zxhlyh in #949
- Feat/model providers by @zxhlyh in #951
- fix: xinference chat completion error by @takatost in #952
- chore: create btn to first place by @iamjoel in #953
- chore: obsolete info api use new api by @iamjoel in #954
- fix: spark provider field name by @zxhlyh in #955
- fix: embedding init err by @takatost in #956
- update document segment by @JohnJyong in #950
- fix: web reader tool retrieve content empty by @takatost in #957
- Feat/add xinference openllm provider by @zxhlyh in #958
- feat: bump version to 0.3.15 by @takatost in #959
- fix: user activation request 404 issue by @MatriQ in #963
- update doc by @crazywoola in #965
- fix: activation page reload issue after activating by @MatriQ in #964
Full Changelog: 0.3.14...0.3.15