LangChain v0.2 Introduces Versioned Docs and New Features for LangSmith
Felix Pinkston Jun 01, 2024 19:34
LangChain v0.2 rolls out versioned docs, LangSmith adds evaluator prompts, dataset splits, and more.
LangChain has announced the release of version 0.2, featuring significant updates to its documentation and new functionalities for LangSmith, as reported on the LangChain Blog.
Product Updates
Highlighting the latest product updates and news for LangChain, LangSmith, and LangGraph
LangChain
The LangChain v0.2 release has versioned docs, with clearer structure and content.
The updated documentation now includes:
- Tutorials: Step-by-step guides for building specific applications, such as chatbots and RAG apps.
- How-to guides: In-depth instructions for advanced tasks.
- Conceptual guides: Glossary of terms and techniques for new concepts.
- API docs: Detailed technical references.
Check out the new docs in Python and JavaScript.
Multi-modal function calling in JavaScript
The new multi-modal function calling in JavaScript allows users to pass images, video, and audio and receive structured output. This update is available in the @langchain/google-genai and @langchain/google-vertexai packages.
LangSmith
Dataset splits for evaluation and filtering/editing dataset examples
LangSmith has introduced dataset splits, making it easier to evaluate subsets of data. Users can tag examples with different split names, edit and add to splits, and filter on desired criteria.
Repetitions to build confidence in your experiment results
LangSmith now allows users to run multiple repetitions of their experiments to account for variability and build confidence in their results.
Off-the-shelf online evaluator prompts
New off-the-shelf evaluators in LangSmith can flag document relevance, check for hallucinations, evaluate answer helpfulness, and flag specific criteria like toxicity.
Upcoming Events
Meet up with LangChain enthusiasts, employees, and eager AI app builders at IRL events this coming month
June 18 (San Francisco): Berkeley LLM meetup. Attendees can learn about LangGraph and other OSS projects built at UC Berkeley. Sign up here.
June 26 (NYC): LangChain and Elastic NYC meetup. The event will feature lightning talks, networking, and refreshments. Sign up here.
Speak the Lang
Real-life use cases and examples of how folks have used LangChain, LangSmith, or LangGraph to build LLM apps with high-quality and accuracy — even in production
Auto-correction for agents
Reflective loops prompt LLMs to reflect on and critique past actions, improving the quality and success rate of agents.
Self-correcting code assistants with LangGraph and Codestral
Codestral-22B, a code generation model from Mistral, has been used to build a self-corrective code assistant. This setup employs structured output, unit testing, and error feedback to self-correct mistakes.
Multi Agent Flow agent with reviewer auto-correction for Text-to-SQL
FlowiseAI’s Multi Agent Flow features agents collaborating on tasks. For text-to-SQL queries, a SQL Expert agent generates the query, an Executor agent runs it, and a Reviewer agent auto-corrects it.
Financial analysis with agents
Can a GPT4-powered AI agent be a good enough performance attribution analyst?
Bruno Veras De Melo from New York Life Investments explores using LangChain agents to assess investment portfolio success drivers, achieving 93% accuracy rates in performance analysis.
LangGraph expense tracker
Jan Willem Altink's project uses LangGraph to manage expenses, allowing users to send pictures of invoices, structure data, and categorize expenses.
Collaboration & Integrations
LangChain said they hope helping users leverage partner features in the ecosystem by using our integrations.
Generative AI Agents Developer Contest by NVIDIA and LangChain
Participants can create a text-based or multimodal agent using NVIDIA’s LLM stack and LangChain or LangGraph, with GPU prizes available. Enter the contest now until June 17th.
Code interpreter with Azure Container App Dynamic Sessions
LangChain's integration with Dynamic Sessions enables safe and reliable execution of Python code by agents.
Forced tool calls with Claude 3
Anthropic’s forced tool calling allows reliable tool usage in LangChain. Users can upgrade their langchain-anthropic package to try these improvements in Python or JavaScript.
Nomic Embed local support for Python
LangChain now supports a local version of Nomic Embed, optimizing performance and cost by dynamically switching between local and remote inference based on input size and complexity.
See how our 1M+ developers and builders are using LangChain in their day-to-day. Thank you for always helping us build better!
Thinking about getting started with LangChain, but don’t know where to begin? Our community members have you covered.
In the 14-min video below, David (Founder @ Decoder.sh) walks through how to build your first chain starting from the fundamentals — using templates, models, and runnables.
Speaking of videos… check out the tutorial below from Santiago Valdarrama (Co-founder @ Tideily) on how to build a RAG app from scratch. The RAG app is built using LangChain, Pinecone, and OpenAI and lets you ask questions about any YouTube video.
On your RAG journey, you may want to split/chunk documents to preserve content while asking questions — just look to Sudarshan Korala’s (Data Science Basics) tutorial.
If you’re exploring more advanced RAG concepts, you might be more interested in:
And if you’re hungry for more content, we have even more resources to keep you nourished.
More Videos:
More Blogs:
And, GitHub Projects:
How can you follow along with all the Lang Latest? Check out the LangChain blog and YouTube channel for even more product and content updates.
You can also join the discussion in our Discord or get support by emailing us at support@langchain.dev.
Image source: Shutterstock.jpg)