Releases: langchain-ai/langchain
Releases · langchain-ai/langchain
v0.0.132
What's Changed
- replicate cleanup by @hwchase17 in #2394
- add metal retriever by @hwchase17 in #2244
- Harrison/pinecone hybrid by @hwchase17 in #2405
- Fix requests wrapper refactor by @jerwelborn in #2417
- hierarchical planning agent for multi-step queries against larger openapi specs by @jerwelborn in #2170
- Update LlamaCpp parameters by @abetlen in #2411
- Don't create a new Pinecone index if doesn't exist by @AdamGS in #2414
- Harrison/elastic search by @hwchase17 in #2419
- improve
AsyncCallbackManager
by @agola11 in #2410 - fix: elasticsearch by @sergerdn in #2402
- Documentation Error - Typo in Docs - Update custom_mrkl_agent.ipynb by @anabranch in #2437
- fixed typo Teplate -> Template by @researchonly in #2433
- Harrison/tfidf retriever by @hwchase17 in #2440
- bump version to 132 by @hwchase17 in #2441
New Contributors
- @abetlen made their first contribution in #2411
- @AdamGS made their first contribution in #2414
- @anabranch made their first contribution in #2437
- @researchonly made their first contribution in #2433
Full Changelog: v0.0.131...v0.0.132
v0.0.131
What's Changed
- Bump anthropic version by @mikelambert in #2352
- fix: testing in Windows and add missing dev dependency by @sergerdn in #2340
- Fixed a typo Teplate > Template by @videowala in #2348
- Fix spelling and grammar in multi_input_tool.ipynb by @jaredthejellyfish in #2337
- Persist database regardless of notebook or script context by @bkamapantula in #2351
- 🐛 fix: missing kwargs in from_agent_and_tools in dataframe agent by @MohammedAlhajji in #2285
- Harrison/base language model by @hwchase17 in #2357
- Update agent_vectorstore.ipynb by @blackaxe21 in #2358
- Expand requests tool into individual methods for load_tools by @happilyeverafter95 in #2254
- Llama.cpp doc update: fix ipynb path by @ipmsteven in #2364
- Remove extra word in CONTRIBUTING.md by @andrewmelis in #2370
- Fix: Pass along kwargs when creating a sql agent by @zachdj in #2350
- Add Enum for agent types by @Shrined in #2321
- requests wrapper by @hwchase17 in #2367
- fix: tests with Dockerfile by @sergerdn in #2382
- Qdrant update to 1.1.1 & docs polishing by @kacperlukawski in #2388
- Harrison/msg files by @hwchase17 in #2375
- Harrison/sql views by @hwchase17 in #2376
- Harrison/gpt4all by @hwchase17 in #2366
- Revision of "elasticearch" spelling problem by @longgui0318 in #2378
- bump version to 131 by @hwchase17 in #2391
New Contributors
- @videowala made their first contribution in #2348
- @bkamapantula made their first contribution in #2351
- @MohammedAlhajji made their first contribution in #2285
- @blackaxe21 made their first contribution in #2358
- @happilyeverafter95 made their first contribution in #2254
- @ipmsteven made their first contribution in #2364
- @andrewmelis made their first contribution in #2370
- @zachdj made their first contribution in #2350
- @Shrined made their first contribution in #2321
- @longgui0318 made their first contribution in #2378
Full Changelog: v0.0.130...v0.0.131
v0.0.130
What's Changed
- Harrison/base agent by @hwchase17 in #2137
- Add Zilliz example by @fzliu in #2288
- micro docstring typo fix by @galtay in #2308
- Fix typo in documentation: vectorstore-retriever.ipynb by @ambujpawar in #2306
- Modify document by @akmhmgc in #2300
- Minor text correction by @akmhmgc in #2298
- Modify output for wikipedia api wrapper by @akmhmgc in #2287
- chat model example by @hwchase17 in #2310
- Reduce number of documents to return for Pinecone by @kennethleungty in #2299
- Introduces SeleniumURLLoader for JavaScript-Dependent Web Page Data Retrieval by @Ynng in #2291
- feat: category support for SearxSearchWrapper by @AlexW00 in #2271
- fixed index name error found at redis new vector test by @DvirDukhan in #2311
- Fixed a typo in an Error Message of SerpAPI by @sleeper in #2313
- Harrison/llama by @hwchase17 in #2314
- custom agent docs by @hwchase17 in #2342
- bump version to 130 by @hwchase17 in #2343
New Contributors
- @galtay made their first contribution in #2308
- @ambujpawar made their first contribution in #2306
- @kennethleungty made their first contribution in #2299
- @Ynng made their first contribution in #2291
- @AlexW00 made their first contribution in #2271
- @DvirDukhan made their first contribution in #2311
- @sleeper made their first contribution in #2313
Full Changelog: v0.0.129...v0.0.130
v0.0.129
What's Changed
- Fixed the link to promptlayer dashboard by @LaloLalo1999 in #2246
- Add total_cost estimates based on token count for openai by @timothyasp in #2243
- update vectorstore.ipynb by @Franfram in #2239
- Add titles to metadatas in gdrive loader by @patrickstorm in #2260
- Update simple typo in llm_bash md by @FourFortyFour in #2269
- Add wikipedia api example by @akmhmgc in #2267
- Add encoding parameter to TextLoader by @Tiger767 in #2250
- Open a Deeplake dataset in read only mode by @jc-tzn in #2240
- remote retriever by @hwchase17 in #2232
- feat: add Dockerfile to run unit tests in a Docker container by @sergerdn in #2188
- Update apis.md by @oldsj in #2278
- Add ability to pass kwargs to loader classes in
DirectoryLoader
, add ability to modify encoding and BeautifulSoup behaviour inBSHTMLLoader
by @samuelwcm in #2275 - Harrison/sql alchemy by @hwchase17 in #2216
- skip unit tests that fail in Windows by @leo-gan in #2238
- Virat/add param to optionally not refresh ES indices by @virattt in #2233
- Harrison/multiline commands by @hwchase17 in #2280
- bump version to 129 by @hwchase17 in #2281
New Contributors
- @LaloLalo1999 made their first contribution in #2246
- @FourFortyFour made their first contribution in #2269
- @akmhmgc made their first contribution in #2267
- @Tiger767 made their first contribution in #2250
- @jc-tzn made their first contribution in #2240
- @oldsj made their first contribution in #2278
- @samuelwcm made their first contribution in #2275
- @leo-gan made their first contribution in #2238
- @virattt made their first contribution in #2233
Full Changelog: v0.0.128...v0.0.129
v0.0.128
What's Changed
- Harrison/docs reqs by @hwchase17 in #2199
- embedding docs by @hwchase17 in #2200
- make requests more general by @hwchase17 in #2209
- Update huggingface_length_function.ipynb by @eltociear in #2203
- feat: document loader for epublications by @MthwRobinson in #2202
- sitemap more flexible by @hwchase17 in #2214
- Add MMR methods to chroma by @Arttii in #2148
- Update vectorstore-retriever.ipynb by @stakodiak in #2210
- Harrison/apify by @hwchase17 in #2215
- add error message to the google drive document loader by @kevinNejad in #2186
- Add
__version__
by @philschmid in #2221 - bump version to 128 by @hwchase17 in #2236
New Contributors
- @stakodiak made their first contribution in #2210
- @kevinNejad made their first contribution in #2186
Full Changelog: v0.0.127...v0.0.128
v0.0.127
What's Changed
- fix memory docs by @hwchase17 in #2157
- Add
temperature
to ChatOpenAI by @shotat in #2152 - Fix docstring (#2147) by @agola11 in #2160
- docs: update docstrings in ElasticVectorSearch class by @sergerdn in #2141
- remove prep by @hwchase17 in #2136
- make tool verbosity override agent verbosity by @agola11 in #2173
- doc: more details on BaseOutputParser docstrings by @blob42 in #2171
- Bugfix: Redis
lrange()
retrieves records in opposite order of inseerting by @zoltan-fedor in #2167 - Harrison/base agent without docs by @hwchase17 in #2166
- [Documents] Updated Figma docs and added example by @maxtheman in #2172
- Add apredict_and_parse to LLM by @timothyasp in #2164
- Fix incorrect wikipage summaries by @ruiaf in #2169
- Add kwargs to from_* in PrompTemplate by @timothyasp in #2161
- Harrison/engine args by @hwchase17 in #2177
- missing word 'not' in constitutional prompts by @gururise in #2176
- Harrison/aim integration by @hwchase17 in #2178
- Harrison/clear ml by @hwchase17 in #2179
- Harrison/arize by @hwchase17 in #2180
- supported async retriever by @Code-Hex in #2149
- Fix wikipedia summaries by @ruiaf in #2187
- Update text.py by @Jeru2023 in #2195
- Small typo in custom_agent.ipynb by @adelevie in #2194
- fix typo in chat_prompt_template docs by @czue in #2193
- fix typing for LLMMathChain by @benheckmann in #2183
- bump version to 127 by @hwchase17 in #2197
New Contributors
- @shotat made their first contribution in #2152
- @sergerdn made their first contribution in #2141
- @maxtheman made their first contribution in #2172
- @ruiaf made their first contribution in #2169
- @gururise made their first contribution in #2176
- @Code-Hex made their first contribution in #2149
- @Jeru2023 made their first contribution in #2195
- @adelevie made their first contribution in #2194
- @czue made their first contribution in #2193
- @benheckmann made their first contribution in #2183
Full Changelog: v0.0.126...v0.0.127
v0.0.126
What's Changed
- add missing
source
field to pymupdf output by @timothyasp in #2110 - Harrison/unstructured validation by @hwchase17 in #2111
- Harrison/better async by @hwchase17 in #2112
- fix typo in docs by @Franfram in #2115
- Fixed 1 missing line in getting_started.md by @AlexTelon in #2107
- add token reduction to ConversationalRetrievalChain by @nkov in #2075
- Harrison/aleph alpha embeddings by @hwchase17 in #2117
- Improve ConversationKGMemory and its function load_memory_variables by @misrasaurabh1 in #1999
- Fix issue#1712: Update
BaseQAWithSourcesChain
to handle space & newline afterSOURCES:
by @walterbm in #2118 - Harrison/gitbook by @hwchase17 in #2044
- Harrison/memory check by @hwchase17 in #2119
- add async support for anthropic by @agola11 in #2114
- searx: implement async + helper tool providing json results by @blob42 in #2129
- fix link by @hwchase17 in #2123
- Strip sitemap entries by @LeSphax in #2132
- Add successful request count to OpenAI callback by @jonpage in #2128
- docs: tiny fix on docs verbiage by @0xcadams in #2124
- Add ability for GoogleDrive loader to load google sheets by @patrickstorm in #2135
- Harrison/memory base by @hwchase17 in #2122
- bump version to 126 by @hwchase17 in #2155
New Contributors
- @Franfram made their first contribution in #2115
- @AlexTelon made their first contribution in #2107
- @nkov made their first contribution in #2075
- @walterbm made their first contribution in #2118
- @LeSphax made their first contribution in #2132
- @jonpage made their first contribution in #2128
- @0xcadams made their first contribution in #2124
- @patrickstorm made their first contribution in #2135
Full Changelog: v0.0.125...v0.0.126
v0.0.125
What's Changed
- Add replicate take 2 by @cbh123 in #2077
- Add OpenWeatherMap API Tool by @Honkware in #2083
- fix(llms): update default AI21 model to j2, as j1 being deprecated by @michael-go in #2108
- v125 by @hwchase17 in #2109
New Contributors
Full Changelog: v0.0.124...v0.0.125
v0.0.124
What's Changed
- big docs refactor by @hwchase17 in #1978
- fix docs by @hwchase17 in #2017
- small nit on index page by @agola11 in #2018
- fix doc tags by @hwchase17 in #2019
- fix headers by @hwchase17 in #2039
- Harrison/open search kwargs by @hwchase17 in #2040
- Harrison/num token calculation by @hwchase17 in #2041
- Added Azure Blob Storage File and Container Loader by @eavanvalkenburg in #1890
- Update getting_started with better example by @jasondotparse in #1910
- fix Poetry 1.4.0+ installation by @dlqqq in #1935
- feat #1915 support for google custom search site restricted api by @gotokatsuya in #1920
- Fix the parameter error of 'Qdrant.maximal_marginal_relevance' by @wy-z in #1921
- Add support for YAML Spec Plugins by @hinthornw in #2054
- PydanticOutputParser unit test by @danielchalef in #2047
- UnstructuredURLLoader: allow url failures, keep processing by @cragwolfe in #1954
- Harrison/redis id key by @hwchase17 in #2057
- Added
.as_retriever()
tofrom_llm()
calls by @Krulknul in #2051 - Removed duplicate "revision_request" in constitutional_ai/prompts.py by @jaredthejellyfish in #2046
- similarity_search is not accepting filters by @rajat1saxena in #1964
- feat: add function similarity_search_limit_score to vectorstores.redis by @hx23840 in #1950
- Updating PromptLayer request in PromptLayer Models to be async in agenerate by @Jped in #2058
- Correctly pass filter down to the similarity_search_with_score function for chroma filtering logic by @Arttii in #1934
- Harrison/site map by @hwchase17 in #2061
- Harrison/document cleanup by @hwchase17 in #2062
- Solve small bug in the kg prompt by @fpingham in #1988
- Improve chat tool prompt by @fpingham in #1989
- enable streaming in anthropic llm wrapper by @agola11 in #2065
- redis retriever by @hwchase17 in #2060
- Harrison/duckdb by @hwchase17 in #2064
- fix: numerous broken documentation links by @alexeldeib in #2070
- [agent_executor] convenience func: lookup tool by name by @blob42 in #2001
- Harrison/fix save context by @hwchase17 in #2082
- Harrison/ai plugin by @hwchase17 in #2084
- docs(deployment): add langchain-serve by @deepankarm in #2006
- docs: fix promptlayer link typo by @michael-go in #2005
- Add prompt and completion token tracking by @iocuydi in #2080
- Harrison/whatsapp loader by @hwchase17 in #2085
- fix(sql_database): related to #2020 by @wqj97 in #2021
- include the tool name for on_tool_end callback by @blob42 in #2000
- notebook for tokens by @hwchase17 in #2086
- Harrison/llama index loader by @hwchase17 in #2097
- feat: Add Notion database document loader by @sbusso in #2056
- Fix issue#1645: Parse llm_output even there's newline (#2092) by @hwchase17 in #2099
- add ConversationalChatAgent to agent.init (fix #2093) by @klein-t in #2098
- Harrison/jina by @hwchase17 in #2043
- Harrison/big query by @hwchase17 in #2100
- rm pandas dependency by @hwchase17 in #2102
- bump version 124 by @hwchase17 in #2101
New Contributors
- @eavanvalkenburg made their first contribution in #1890
- @jasondotparse made their first contribution in #1910
- @dlqqq made their first contribution in #1935
- @gotokatsuya made their first contribution in #1920
- @wy-z made their first contribution in #1921
- @Krulknul made their first contribution in #2051
- @jaredthejellyfish made their first contribution in #2046
- @rajat1saxena made their first contribution in #1964
- @hx23840 made their first contribution in #1950
- @Arttii made their first contribution in #1934
- @alexeldeib made their first contribution in #2070
- @deepankarm made their first contribution in #2006
- @michael-go made their first contribution in #2005
- @wqj97 made their first contribution in #2021
- @sbusso made their first contribution in #2056
Full Changelog: v0.0.123...v0.0.124
v0.0.123
What's Changed
- (ChatOpenAI) Add model_name to LLMResult.llm_output by @mariokostelac in #1960
- plugin tool by @hwchase17 in #1974
Full Changelog: v0.0.122...v0.0.123