great question! yes – we have built in tracing, including support for a locally run docker image https://
langchain.readthedocs.io/en/latest/trac
ing.html
…
@langchain
-
LangChain Adds Built-in Tracing Support Docker Image
By
–
-
LangChain Integration with BERTopic Now Available
By
–
Really excited to see LangChain integrated with BERTopic! Thanks @MaartenGr https://
maartengr.github.io/BERTopic/getti
ng_started/representation/representation.html#langchain
… -
Companies Enabling OpenAI Call Logging and Tracking Integration
By
–
There a bunch of awesome companies helping with logging/tracking of calls to OpenAI. We're excited to start integrating with them! Helicone (
@justinstorre
): https://
langchain.readthedocs.io/en/latest/ecos
ystem/helicone.html
… PromptLayer (
@imjaredz
): https://
langchain.readthedocs.io/en/latest/ecos
ystem/promptlayer.html
… -
Chroma Persistence and Agent Refactors Update
By
–
Other changes include: Chroma (
@trychroma
) persistance – @atroyn Standardize prompt loading – Ibis Prevedello
Standardize makefile – blob42
Added to retries – @klingefjord Agent refactors – Amos Ng Doc Fixes: @mm_sasmitha @dhruv___anand -
Open Source Model Integrations and LLM Platform Updates
By
–
v0.0.86 Lots more open source model integrations! @EnricoShippole PromptLayer (
@imjaredz
) and Helicone (
@justinstorre
) integrations And lots of other docs and bug fixes! -
Using Multiple LLMs Together in Chain Workflows
By
–
Yup! You can use them both as different LLMs in a chain
-
LangChain Chroma Vector Store Integration Announcement
By
–
Really excited to announce a new vector store integration: @trychroma If you wanted to develop easily with local vector stores, but ran into issues with FAISS… you should try this Blog Post: https://
blog.langchain.dev/langchain-chro
ma/
… Examples Repo: -
Graph Memory for Chatbot Knowledge Extraction
By
–
Graph Memory Use this graph data structure as a memory module in your chatbot. Extracts knowledge triplets every interaction, builds a graph over time, and pulls relevant triplets into context when replying Documentation: https://
langchain.readthedocs.io/en/latest/modu
les/memory/getting_started.html#conversation-knowledge-graph-memory
… -
LangChain Adds Graph Support for Question Answering
By
–
Graph Support Previously in LangChain, the only real data structure used for storing data was a vector store In v0.0.85 we've now added in [BETA] graph support This can be used for Question Answering and Memory
-
Graph Question-Answering with Knowledge Triplets Extraction
By
–
Graph Question-Answering Create a graph by extracting knowledge triplets. Then for a given question extract the relevant entities and pull in information about them Documentation: https://
langchain.readthedocs.io/en/latest/modu
les/chains/combine_docs_examples/graph_qa.html
…