AI BriefingLangChain Blog news
The latest AI and developer news from LangChain Blog, with short summaries.
Visit website →A Perspective on Agent Frameworks
Building reliable agentic systems requires understanding the difference between workflows and agents, and controlling the appropriate context.
LangSmith Introduces End-to-End OpenTelemetry Support
LangSmith now supports OpenTelemetry, standardizing tracing for LLM applications and providing unified observability.
How to Quickly Get Started with LLM Evaluation Using OpenEvals
LangChain has released OpenEvals, an open-source framework that helps evaluate the performance of LLM and agent applications.
Benchmarking Single-Agent Performance
The study analyzed how the number of instructions and tools provided to a ReAct agent affects its performance.
LangChain 2024 State of AI Report
LangChain has released its annual report covering 2024 AI development trends and LLM infrastructure usage.
LangSmith Begins Supporting OpenTelemetry
LangSmith now supports the OpenTelemetry format, strengthening its distributed tracing and observability capabilities.
Promptim: An Experimental Library for Prompt Optimization
The LangChain team has released the Promptim library, which automatically improves prompts based on datasets and evaluation metrics.
Improving Memory Retrieval Performance: New Computer Achieves 50% Recall Boost with LangSmith
New Computer leveraged LangSmith to improve its agent memory system's recall by 50% and precision by 40%.
Pairwise Evaluation Using LangSmith
LangSmith has launched a Pairwise Evaluation feature that directly compares which of two LLM responses is superior, enabling more accurate measurement of human preference.
Regression Testing with LangSmith
LangSmith has launched improved regression testing features that allow precise comparison and analysis of performance changes in LLM applications.
LangSmith Launches as a Directly Purchasable Service on Azure Marketplace
LangSmith, LangChain's LLM DevOps platform, has launched as an Azure Kubernetes Application through Azure Marketplace.
Benchmarking Query Analysis Under High Cardinality Conditions
We benchmark the problem of handling high-cardinality categorical values that arises during LLM-based query analysis, and its solutions.
Ensuring LLM Reliability Through Evaluation Driven Development (EDD)
Dosu adopted the Evaluation Driven Development (EDD) approach to prevent performance degradation in LLM products and enhance reliability.
Building and Deploying a RAG App Using Pinecone Serverless
Introduces how to build and deploy a production-grade RAG application by combining Pinecone Serverless, LangServe, and LangSmith.
LangChain 2023 State of AI Report
LangChain analyzed LangSmith data to reveal 2023 trends in LLM application development and testing.
Benchmarking Agent Tool-Use Capabilities
LangChain has released 4 new testing environments that measure the tool use and function calling performance of LLM agents.
Data Extraction Benchmarking
The performance of GPT-4, Claude, and open-source LLMs was compared and analyzed when extracting structured data from chat logs.
LangSmith Unveils Benchmark Sharing Feature
LangChain has launched a dataset and results sharing feature in LangSmith to help developers easily compare and validate LLM architectures.
Test Run Comparison Feature
LangChain unveiled a new LangSmith feature that lets you compare multiple test run results side by side for evaluating LLM applications.
Testing Fine-Tuned Open Source Models on LangSmith
This introduces a process for systematically evaluating and comparing the performance of fine-tuned open source models using LangSmith.