Releases: langgenius/dify
v0.4.4
Add Together.ai model provider and fix some bugs.
What's Changed
- feat: optimize memory & invoke error output by @takatost in #1931
- fix: vision fail in complete app by @takatost in #1933
- Feat/optimize entity construct by @takatost in #1935
- feat: reuse decoding_rsa_key & decoding_cipher_rsa & optimize construct by @takatost in #1937
- fix: llm result usage none by @takatost in #1940
- fix: gpt-4-32k model name empty in OpenAI response by @takatost in #1941
- fix: arg user required and must not be null in service generate api by @takatost in #1943
- fix: text completion app cannot get data. by @GarfieldDai in #1942
- fix: text-generation run batch by @zxhlyh in #1945
- fix streaming by @guchenhe in #1944
- Add Together.ai's OpenAI API-compatible inference endpoints by @guchenhe in #1947
- feat: reorder togetherai by @takatost in #1951
- change API key field to 'required' by @guchenhe in #1953
- optimize qa prompt by @JohnJyong in #1957
- prohibit enable and disable function when segment is not completed by @JohnJyong in #1954
- delete remnant 'required': false by @guchenhe in #1955
- fix: check result should be string. by @GarfieldDai in #1959
- fix: generate not stop when pressing stop link by @takatost in #1961
- bump version to 0.4.4 by @takatost in #1962
Full Changelog: 0.4.3...0.4.4
v0.4.3
Optimize performance & fix few bugs.
What's Changed
- fix: zhipuai chatglm turbo prompts must user, assistant in sequence by @takatost in #1899
- fix: customize model schema fetch failed raise error by @takatost in #1900
- Fix/add qdrant timeout default value by @JohnJyong in #1901
- fix: remove useless code by @zxhlyh in #1903
- fix: read file encoding error by @newsouther in #1902
- fix: tongyi generate error by @takatost in #1904
- fix: array oob in azure openai embeddings by @crazywoola in #1905
- fix: spark credentials validate by @takatost in #1906
- fix: notion_indexing_estimate embedding_model_instance NPE by @takatost in #1907
- fix: window size changed causes result regeneration by @iamjoel in #1908
- fix: model quota format by @zxhlyh in #1909
- fix: remove useless code. by @GarfieldDai in #1913
- ci: Properly cache pip packages by @bowenliang123 in #1912
- web: Add style CI workflow to enforce eslint checks on web module by @bowenliang123 in #1910
- fix customer spliter character by @JohnJyong in #1915
- feat: model parameter prefefined by @zxhlyh in #1917
- fix: email template style by @takatost in #1914
- Fix comparison bug in ApplicationQueueManager by @waltcow in #1919
- fix: app create raise error when no available model providers by @takatost in #1921
- feat: optimize invoke errors by @takatost in #1922
- alter schedule timedelta by @JohnJyong in #1923
- fix a typo in file agent_app_runner.py by @liupeng0913 in #1927
- fix: xinference cache by @Yeuoly in #1926
- feat: optimize performance by @takatost in #1928
- feat: bump version 0.4.3 by @takatost in #1930
New Contributors
- @newsouther made their first contribution in #1902
- @liupeng0913 made their first contribution in #1927
Full Changelog: 0.4.2...0.4.3
v0.4.2
What's Changed
- fix: ciphertext error by @takatost in #1872
- Fix/model runtime by @Yeuoly in #1873
- fix: openai origin credential not start with { by @takatost in #1874
- fix: more like this. by @GarfieldDai in #1875
- fix: remove tiktoken from text splitter by @Yeuoly in #1876
- doc: Respect and prevent updating existed yarn lockfile when installing dependencies by @bowenliang123 in #1871
- fix: openai embedding list out of bound by @crazywoola in #1879
- fix: model parameter modal initial value by @zxhlyh in #1883
- fix: model parameter stop sequence by @zxhlyh in #1885
- Fix variable name in AgentApplicationRunner by @waltcow in #1884
- fix: gemini block error by @takatost in #1877
- fix: huggingface and replicate. by @GarfieldDai in #1888
- fix: anthropic messages empty raise errors by @takatost in #1893
- Grammar fix by @BorisPolonsky in #1892
- add qdrant client timeout limit by @JohnJyong in #1894
- add example api url endpoint in placeholder by @guchenhe in #1887
- fix: zhipuai history format wrong by @takatost in #1897
- bump version to 0.4.2 by @takatost in #1898
New Contributors
- @BorisPolonsky made their first contribution in #1892
Full Changelog: 0.4.1...0.4.2
v0.4.1
What's Changed
- fix: agent strategy missing in app model config by @takatost in #1863
- fix: default model parameter precision by @takatost in #1864
- fix: hit testing throws errors cause internal server error by @takatost in #1865
- fix: provider create cause IntegrityError by @takatost in #1866
- fix: Azure text-davinci-003 missing by @takatost in #1867
- fix: IntegrityError import wrong by @takatost in #1868
- fix: xinference secret server_url by @Yeuoly in #1869
- bump version to 0.4.1 by @takatost in #1870
Full Changelog: 0.4.0...0.4.1
v0.4.0
🎉🎉 Dify's Version 0.4 is out now.
We've made some serious under-the-hood changes to how the Model Runtime works, making it more straightforward for our specific needs, and paving the way for smoother model expansions and more robust production use.
What's Changed
-
Model Runtime Rework: We've moved away from LangChain, simplifying the model layer. Now, expanding models is as easy as setting up the model provider in the backend with a bit of YAML.
For more details, see: https://github.com/langgenius/dify/blob/main/api/core/model_runtime/README.md
-
App Generation Update: Replacing the old
Redis Pubsub
queue withthreading.Queue
for a more reliable, performant, and straightforward workflow. -
Model Providers Upgraded: Support for both preset and custom models, ideal for adding OpenAI
fine-tuned models
or fitting into various MaaS platforms. Plus, you can now check out supported models without any initial configuration. -
Context Size Definition: Introduced distinct
context size
settings, separate fromMax Tokens
, to handle the different limits and sizes in models like OpenAI'sGPT-4 Turbo
. -
Flexible Model Parameters: Customize your model's behavior with easily adjustable parameters through YAML.
-
GPT-2 Tokenizer Files: Now cached within Dify's codebase, making builds quicker and solving issues related to acquiring tokenizer files in offline source deployments.
-
Model List Display: The App now displays all supported preset models, including details on any that aren't available and how to configure them.
-
New Model Additions: Including Google's
Gemini Pro
andGemini Pro Vision
models (Vision requires an image input), Azure OpenAI'sGPT-4V
, and support forOpenAI-API-compatible
providers. -
Expanded Inference Support:
Xorbit Inference
now includes chat mode models, and there's a wider range of models supporting Agent inference. -
Updates & Fixes: We've updated other model providers to be in sync with the latest version APIs and features, and squashed a series of minor bugs for a smoother experience.
Catch you in the code,
The Dify Team 🛠️
Change Log
- fix: not return annotation author error happens by @iamjoel in #1793
- fix typo by @kztao in #1791
- Compatible with the situation where there is no user information. by @JohnJyong in #1792
- ppt & pptx improve by @JohnJyong in #1790
- Fix: delete member dropdown not shown by @crazywoola in #1794
- fix: settings/members dropdown ui by @crazywoola in #1797
- fix gpt 4v upload image issue by @JohnJyong in #1799
- feat: add billing switch. by @GarfieldDai in #1789
- Update README.md by @joshua20231026 in #1802
- Update README.md by @joshua20231026 in #1803
- Update README_ES.md by @joshua20231026 in #1807
- Update README_JA.md by @joshua20231026 in #1805
- Update README_CN.md by @joshua20231026 in #1806
- Update README_CN.md by @joshua20231026 in #1804
- feat: add email template for invite new user in workspace by @crazywoola in #1810
- Fix azure openai gpt4v&1106 config by @charli117 in #1811
- fix: mail link color by @crazywoola in #1812
- Parse base64 eml file by @charli117 in #1796
- add link to canny to README, plus some rewording by @guchenhe in #1814
- feat: add roadmap and feedback link by @iamjoel in #1816
- fix unstructured requirements by @JohnJyong in #1821
- Fix Azure OpenAI Provider BASE_MODELS by @zhangao0086 in #1813
- fix: edited by is missing by @crazywoola in #1824
- Update README.md by @crazywoola in #1825
- fix: selection too long break ui by @iamjoel in #1826
- Fix chitchat lost context by @charli117 in #1828
- feat: nginx add "restart: always" by @takatost in #1829
- feat: comment db port to host by @takatost in #1831
- fix file type not support when preview by @JohnJyong in #1841
- fix segment update issue by @JohnJyong in #1844
- fix: splitting text ui broken by @iamjoel in #1848
- Feat/add community link to dropdown by @crazywoola in #1851
- ci: Bump Docker Github actions by @bowenliang123 in #1852
- Refactor BaseVectorIndex delete method by @waltcow in #1853
- Add data clean schedule by @JohnJyong in #1859
- fix file estimate issue by @JohnJyong in #1860
- Model Runtime by @takatost in #1858
- frontend for model runtime by @zxhlyh in #1861
New Contributors
- @joshua20231026 made their first contribution in #1802
- @zhangao0086 made their first contribution in #1813
- @bowenliang123 made their first contribution in #1852
Full Changelog: 0.3.34...0.4.0
v0.3.34
Features
- Annotation Reply, see details: Link
- Dify Knowledge supports
unstructured.io
as the file extraction solution. - Azure OpenAI add
gpt-4-1106-preview
、gpt-4-vision-preview
models support. - SaaS services now support replacing the logo of WebApp after subscribing.
Important Upgrade Notice
-
Annotation Reply
The annotation function can support direct replies to related questions, so we need to assign values to previously unstored questions for the table
message_annotations
.we need doing below command in your api docker container
docker exec -it docker-api-1 bash flask add-annotation-question-field-value
or direct run below command when you launch from source codes.
cd api flask add-annotation-question-field-value
-
Unstructured.io Support
Due to the support of this feature, we have added four new formats of text parsing( msg , eml, ppt, pptx ) and optimized two text parsing formats (text, markdown) in our SAAS erviroment.
For localhost you need to do below actions to support unstructured.io
docker pull
from unstructured's image repository.
docker pull downloads.unstructured.io/unstructured-io/unstructured-api:latest
- Once pulled, you can launch the container
docker run -d --rm --name unstructured-api downloads.unstructured.io/unstructured-io/unstructured-api:latest --port 8000 --host 0.0.0.0
- In our docker-compose.yaml, add two new environment variables for the
api
andworker
services.
ETL_TYPE=Unstructured UNSTRUCTURED_API_URL=http://unstructured:8000/general/v0/general
- Restart the Dify‘s services
docker-compose up -d
What's Changed
- feat: Annotation management by @JohnJyong in #1767
- feat: annotation management frontend by @iamjoel in #1764
- feat: add unstructured.io support by @JohnJyong in #1780
- feat: unstructured frontend by @iamjoel in #1777
- feat: azure openai add gpt-4-1106-preview、gpt-4-vision-preview models by @charli117 in #1751
- feat: custom webapp logo by @zxhlyh in #1766
- feat: can replace logo. by @GarfieldDai in #1752
- feat: web app support some feature by @zxhlyh in #1753
- feat: update pricing by @iamjoel in #1749
- feat: remove billing config. by @GarfieldDai in #1761
- feat: only tenant owner can subscription. by @GarfieldDai in #1770
- feat: add self checks to issues and discussions templates by @guchenhe in #1742
- feat: remove stripe and anthropic. by @GarfieldDai in #1746
- feat: bump version to 0.3.34 by @takatost in #1788
- chore: update term links by @iamjoel in #1748
- chore: update annotation reply english i18n by @iamjoel in #1773
- chore: enchance annotation ui by @iamjoel in #1781
- chore: enchance ext name by @iamjoel in #1787
- chore: update bm25 search properties by @ethuwlwfu3288 in #1758
- fix: default types by @crazywoola in #1728
- fix: prompt null parse var error by @iamjoel in #1731
- fix: server component use translate errorts lint error by @iamjoel in #1732
- fix only full text search in app issue by @JohnJyong in #1736
- fix: qa regex by @crazywoola in #1738
- fix: check rerank document is not empty by @JohnJyong in #1740
- fix: ad block disabled tracking would block ga then can not pay by @iamjoel in #1741
- fix: issue templates not render correctly by @crazywoola in #1763
- fix annotation reply in universal chat by @JohnJyong in #1768
- fix annotation query exception by @JohnJyong in #1771
- fix: not owner can not pay by @iamjoel in #1772
- fix: get billing info in self-hosted edition from current workspace by @zxhlyh in #1774
- fix: add then eidt annotion cause show bug by @iamjoel in #1775
- fix: vdb index issue by @JohnJyong in #1776
- fix: web app text by @zxhlyh in #1784
- fix: unstructured file extension by @zxhlyh in #1785
New Contributors
- @guchenhe made their first contribution in #1742
- @ethuwlwfu3288 made their first contribution in #1758
Full Changelog: 0.3.33...0.3.34
v0.3.33
What's Changed
- feat: add jina embedding by @zxhlyh in #1647
- feat: fe mobile responsive next by @yuhao1118 in #1609
- feat: clipboard paste by @yuhao1118 in #1663
- feat: regenrate conversation in embeded window by @crazywoola in #1708
- feat: drag to upload image by @yuhao1118 in #1666
- feat: add items by @crazywoola in #1633
- feat:billing by @GarfieldDai in #1679
- feat: price plan frontend by @iamjoel in #1683
- feat: remove documents limit by @zxhlyh in #1697
- feat: billing enhancement 20231204 by @GarfieldDai in #1691
- feat: add upgrade ga test by @iamjoel in #1690
- feat: use gtag instead gtm by @iamjoel in #1695
- refactor: typo in dataset docstore by @perzeuss in #1711
- Remove legacy docker startup docs in frontend by @crazywoola in #1645
- chore(deps): bump semver from 5.7.1 to 5.7.2 in /web by @dependabot in #1680
- chore(deps): bump word-wrap from 1.2.3 to 1.2.5 in /web by @dependabot in #1681
- chore(docker-compose): use proper comment indentation for users to easily toggle on and off features by @nekomeowww in #1648
- Chore: change dataset's i18n to knowledge by @iamjoel in #1629
- Update README.md by @Panmuse in #1636
- Update README.md by @Panmuse in #1623
- doc: use proper links by @crazywoola in #1673
- fix score_threshold_enabled name by @JohnJyong in #1626
- fix: score_threshold_enabled variable by @zxhlyh in #1627
- fix: old webapp url still valid by @crazywoola in #1643
- fix: batch mobile layout fixes by @yuhao1118 in #1641
- fix #1637 by @wangbooth in #1638
- fix: app card hover selector by @yuhao1118 in #1646
- fix: Jina AI logo by @zxhlyh in #1656
- fix: explore page header menu hide in safari by @iamjoel in #1658
- fix: documents index mismatch error in rerank by @wangbooth in #1662
- fix: app setting click pop by @kimjion in #1660
- fix: Fix typo in documentation: change 'converation' to 'conversation' by @simpx in #1665
- fix: Incorrect order of embedded documents in CacheEmbedding by @ewen-ai in #1671
- fix wrong syntax of type definitions by @Yeuoly in #1678
- fix: vector_size convert bytes to MB. by @GarfieldDai in #1684
- fix(web): reserve default copy behavior by @JunIce in #1693
- fix: error Class 'builtins.list' is not mapped by @perzeuss in #1710
- fix: process document priority tip by @zxhlyh in #1712
- fix ascii codec error, by using utf8 by @yunyu-Mr in #1608
- fix: auto generate type error in controllers/web/conversation.py by @crazywoola in #1718
- fix: remove postgresql default charset config by @takatost in #1720
- bump version to 0.3.33 by @takatost in #1719
New Contributors
- @nekomeowww made their first contribution in #1648
- @kimjion made their first contribution in #1660
- @simpx made their first contribution in #1665
- @ewen-ai made their first contribution in #1671
- @Yeuoly made their first contribution in #1678
- @perzeuss made their first contribution in #1710
- @yunyu-Mr made their first contribution in #1608
Full Changelog: 0.3.32...0.3.33
v0.3.32
New Features
- Supported the
rerank
model ofxinference
for local deployment, such as:bge-reranker-large
,bge-reranker-base
. - ChatGLM2/3 support
⚠️ Breaking Change
We've recently switched provider ChatGLM
to the OpenAI API protocol, so from now on, we'll only be supporting ChatGLM3
and ChatGLM2
. Unfortunately, support for ChatGLM1
has been deprecated.
What's Changed
- fix: retrieval test page hide rerank model also hide retrieval config by @iamjoel in #1585
- fix: chat on start bug by @iamjoel in #1588
- Update docker-compose.yaml by @JohnJyong in #1587
- feat: bump version to 0.3.31-fix1 by @takatost in #1589
- fix: safari can not in by @iamjoel in #1590
- feat: add anthropic claude-2.1 support by @takatost in #1591
- feat: bump version to 0.3.31-fix2 by @takatost in #1592
- fix sql transaction error in statistic API by @waltcow in #1586
- fix: chat token spent info style by @zxhlyh in #1597
- fix weaviate hybrid search issue by @JohnJyong in #1600
- feat: optimize db connections in thread by @takatost in #1601
- fix: hybrid search when document is none by @JohnJyong in #1603
- Fix/multi thread parameter by @JohnJyong in #1604
- feat: bump version to 0.3.31-fix3 by @takatost in #1606
- feat: xinference rerank model support by @takatost in #1615
- feat: chatglm3 support by @takatost in #1616
- fix: dataset tool missing in n-to-1 retrieve mode by @takatost in #1617
- fix: provider chatglm tests error by @takatost in #1618
- feat: add xinference rerank model by @zxhlyh in #1619
- feat: bump version to 0.3.32 by @takatost in #1620
New Contributors
Full Changelog: 0.3.31...0.3.32
v0.3.31-fix3
Major Fixes
- Fix: Hybrid search recall fails when the chosen dataset contains no valid documents.
Important Notice
Before upgrading to 0.3.31
, please ensure that all steps outlined in the Important Upgrade Notice for the 0.3.31 release are completed: https://github.com/langgenius/dify/releases/tag/0.3.31
What's Changed
- fix sql transaction error in statistic API by @waltcow in #1586
- fix: chat token spent info style by @zxhlyh in #1597
- fix weaviate hybrid search issue by @JohnJyong in #1600
- feat: optimize db connections in thread by @takatost in #1601
- fix: hybrid search when document is none by @JohnJyong in #1603
- Fix/multi thread parameter by @JohnJyong in #1604
- feat: bump version to 0.3.31-fix3 by @takatost in #1606
New Contributors
Full Changelog: 0.3.31-fix2...0.3.31-fix3
v0.3.31-fix2
New Features
- add anthropic
claude-2.1
support
Major Fixes
- fix: white page problem in Safari.
Important Notice
Before upgrading to 0.3.31
, please ensure that all steps outlined in the Important Upgrade Notice for the 0.3.31 release are completed: https://github.com/langgenius/dify/releases/tag/0.3.31
What's Changed
- fix: safari can not in by @iamjoel in #1590
- feat: add anthropic claude-2.1 support by @takatost in #1591
- feat: bump version to 0.3.31-fix2 by @takatost in #1592
Full Changelog: 0.3.31-fix1...0.3.31-fix2