Parallel function calling for agents With OpenAI's new models, it's now easier than ever to do have agents call tools in parallel Why is this a big improvement? This enables fewer calls to the language model, saving on time and money Check it out in action as it
@langchain
-
Advanced RAG Templates with LangChain for OpenAI Assistants
By
–
Advanced retrieval templates General consensus seems to be that the built-in retrieval in OpenAI assistants is fine for basic/small use cases, but for more complex ones you are going to want to more control This is where LangChain templates for advanced RAG comes in. Some of
-
LangSmith Adds Multi-Modal Image Tracing for LLM Debugging
By
–
Multi-modal LangSmith Tracing With OpenAI releasing GPT4-V, we've now adding support for images in LangSmith tracing! This means you can now see images as they appear in your LLM calls, making it easy to debug multi-modal RAG pipelines Example Trace:
-
Parallel Function Calling Improves Data Extraction Significantly
By
–
Parallel function calling for extraction The newest OpenAI models allow you do parallel function calling The main focus here as been on how this improves assistants, but this also makes extraction **significantly** easier for the following reasons: Less logic needed to
-
Context by Fleet: Python Ecosystem Documentation Embeddings
By
–
Context by Fleet is an awesome project to embed the docs of basically the entire Python ecosystem. Use it to give your LLM up-to-date information about just about any library.
— LangChain (@LangChain) 8 novembre 2023
Check out how you can build your own Chat LangChain experience using Context's embeddings of the… https://t.co/ElmZog6ftbContext by Fleet is an awesome project to embed the docs of basically the entire Python ecosystem. Use it to give your LLM up-to-date information about just about any library. Check out how you can build your own Chat LangChain experience using Context's embeddings of the
-
SPADE: Automated Test Generation from LLM Prompt Refinements
By
–
Sick of debugging finicky LLM apps? 😫
— LangChain (@LangChain) 8 novembre 2023
Introducing ♠️ SPADE, a tool that automatically generates tests from prompt refinements! 🧪
Paste any prompt from the LangChain Hub and it will tailor eval recommendations using prompt deltas. 🔁
SPADE analyzes version history to suggest… pic.twitter.com/eSmWaPjqc8Sick of debugging finicky LLM apps? Introducing SPADE, a tool that automatically generates tests from prompt refinements! Paste any prompt from the LangChain Hub and it will tailor eval recommendations using prompt deltas. SPADE analyzes version history to suggest
-
LangChain Introduces Fallbacks for LLM API Error Handling
By
–
Introducing Fallbacks in It’s not uncommon to encounter issues with LLM API's. In production, you need to gracefully handle such issues. We’ve introduced Fallbacks to the LangChain Expression Language (LCEL) to help with just that. Available in Python and JS! a :
-
LangChain Fallbacks Prevent LLM Outages and Downtime
By
–
LLM Outages?! No problem. Use LangChain fallbacks to avoid any downtime! https://
python.langchain.com/docs/guides/fa
llbacks
… -
OpenAI Assistants API Integration with LangChain Tools
By
–
OpenAI Assistants API Spin up OpenAI assistants and run them as any other LangChain agent! This involves connecting them to LangChain's 100+ tools, OpenAI's code interpreter + retrieval, or easily defining your own! Try it out yourself: https://
github.com/langchain-ai/l
angchain/blob/master/cookbook/openai_v1_cookbook.ipynb
… -
Building Personalized RAG Chatbots with Social Media Data
By
–
Conversational RAG LLM App with Social Media Context Data A great project by Claire Longo shows how to create ClaireBot This chatbot is personalized with her voice and her knowledge (from social media) It's still an open question how to best customize a chatbot with