LangChain

✨ Today, we’re thrilled to announce ✨ - The general availability of LangSmith (no more waitlist!) - Our Series A fundraise led by Sequoia Capital - Our beautiful new homepage and brand We've worked hard over the past few months to add requested features and ensure LangSmith can operate at scale. We’re now confident in saying that it is the most complete platform for building production-grade LLM applications, whether or not you’re using LangChain. Learn more here: https://lnkd.in/gZxW8X_V and sign up here: https://lnkd.in/dwXZt_ZT Our series A round will give us the capital needed to grow our open source and platform offerings. Working with Sonya Huang, Romie Boyd, and the rest of the Sequoia team has been a privilege so far! https://lnkd.in/g8nw36_Z Finally, we’re excited to unveil our new homepage and brand. Dive into our new website at https://www.langchain.com/ to see the changes for yourself, explore the expanded resources, and discover what LangChain, LangSmith, and LangServe have to offer. PS — we’re hiring! Explore our careers page and reach out if you think you’re a fit for any of our open positions! https://lnkd.in/g9rXjrvC

LangChain 🤝 Nvidia NIM APIs Nvidia just launched NIM APIs for a HUGE selection of AI models They are all easily usable within LangChain and are highlighted as one of the recommend ways to use LLM: https://lnkd.in/gVfZ4Dgd Reranking: https://lnkd.in/gAd43Xja Embedding: https://lnkd.in/ghfSRwWC

🧪 Join us Aug. 7 at 10 AM PDT as we talk Synthetic Data for Enterprise! Discover how Synthetic Data Generation (SDG) is accelerating LLM application development for companies. 🚀 📚 Join us live to learn how to: ⚒️ Use LangSmith to quickly test production LLM applications. ⚗️ Generate the right synthetic data for your use case. 📈 Improve performance of your application based on synthetic test data results! We’ll cover a real-world use case in financial services during this event; join us to get up to speed quickly on one of the most important emerging industry-relevant techniques! S/o to LangChain for partnering! RSVP: https://lnkd.in/g7-3ikWp

🔢 Dataset schemas for fast & iterative data curation in LangSmith When building a dataset iteratively for an LLM app, having a defined schema for testing across examples lets you avoid broken code and keep your data clean and consistent. With LangSmith, you can now define and flexibly manage dataset schemas. LangSmith validates your examples against the defined schema, throwing errors when detected. Updating your schema is also simplified in LangSmith — you can go through a queue of datapoints that no longer fit the desired schema, fixing them directly in the UI. ✍️ Read the blog post: https://lnkd.in/gCXTzAqA 📓 See the docs: https://lnkd.in/gCwCYSVa 🌟 Try it out in LangSmith: smith.langchain.com

🤖 Evaluate agents on SWE-Bench SWE-Bench is one of the most popular (and difficult) benchmarks for developers to test coding agents against. In this video, we walk through how to load the SWE-benchmark dataset and evaluate agents on it in LangSmith. LangSmith lets you debug issues quickly and iterate on your agent to improve performance. 📽️ Watch the video: https://lnkd.in/gutXP9R8 📓 Check out the docs: https://lnkd.in/gpSV3GZn

💻 Join us for an Agents and Compound AI Hackathon in San Francisco on Sunday, August 11th, hosted by Fireworks AI, Factory, and LangChain. Apply here ➡ https://lu.ma/kwp4mkr3 What exactly defines an agent? An agent is a system that uses an LLM to determine the control flow of an application. The more an LLM dictates the system's behavior, the more "agentic" it becomes. The autonomous capabilities of agents will enable new, innovative applications. Agents are a fantastic example of compound AI. Compound AI systems don’t just use a single mega model, they combine multiple interacting components, such as retrievers, tools and specialized models. This hackathon is your chance to collaborate, innovate, and push the boundaries of what's possible with agents and compound AI 🤖 ✨

Handle rate limits with ease for any chat model in LangChain. Our built-in, thread-safe rate limiter lets you control request rates to avoid hitting API limits set by your chat model providers. This rate limiter is available as of `langchain-core 0.2.24`. Learn more & see how to initialize and use rate limiting: https://lnkd.in/gaH3AsHq

Unstructured 🤝 LangChain With our latest partner package with unstructured.io, you can easily process a variety of file types into documents, which can be used for vector-store retrieval. The `langchain-unstructured == 0.1.0` package contains a production-ready hosted API from Unstructured, plus their open source local file processing. See the docs: https://lnkd.in/gbkK_TDb

🗺️ Map-reduce operations for parallel execution To parallelize tasks effectively, you can use map-reduce operations to break tasks into sub-tasks, process them simultaneously, and aggregate results. In this video, see how LangGraph's native support for map-reduce operations can handle unknown objects and distribute different states to multiple nodes seamlessly. This allows you to  manage flexible and dynamic workflows. 📽️ Video: https://lnkd.in/g83CFuhV 📓 Docs: https://lnkd.in/guSrPYRH

As agentic applications become more popular, we are seeing longer traces getting sent to LangSmith. Oftentimes, you want to search for a particular event within a single long trace. Our new feature enable exactly this, allowing you to filter for runs within the trace view. Learn more: https://lnkd.in/g_UEfiQm