Version 0.3.15

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