FastChat

FastChatSince 2026-08

FastChat is an open-source platform developed by LMSYS (Large Model Systems Organization) for training, serving, and evaluating large language model-based chatbots. As the technology powering the popular Chatbot Arena leaderboard, FastChat provides state-of-the-art serving infrastructure with OpenAI-compatible REST APIs, distributed worker orchestration, and Web UI interfaces. Machine learning engineers and enterprise developers use FastChat to self-host open-weights models (like Llama 3, Mistral, Vicuna, and DeepSeek) with multi-GPU acceleration and vLLM integration.

FastChat provides an end-to-end stack: high-throughput model serving workers, a central controller for load balancing across GPU nodes, and an OpenAI-compatible API server. It also includes comprehensive fine-tuning recipes using Hugging Face Transformers, DeepSpeed, and FlashAttention-2. FastChat is the gold standard foundation for organizations establishing sovereign, on-premise AI chat and API infrastructure.

Browser UseSince 2026-01

Browser Use is an open-source Python library that connects LLMs to browser automation pipelines, enabling AI agents to navigate websites, interact with dynamic DOM elements, bypass multi-step forms, and extract structured data autonomously. Built on top of Playwright and LangChain, it provides vision-augmented element detection and deterministic state tracking. Unlike traditional headless scrapers, Browser Use feeds DOM tree snapshots and viewport screenshots to multimodal models like Claude 3.7 Sonnet or GPT-4o, allowing agents to understand complex UI layouts, handle popups, solve interactive workflows, and execute sequential tasks in plain English.

Browser Use utilizes an innovative DOM accessibility tree pruning algorithm to minimize token consumption while maintaining complete interactive context. The library exposes clean async Python primitives, custom action handlers, and persistent session cookies, making it ideal for automating authentication-heavy corporate portals, booking workflows, and complex SaaS configurations. The agent logs every click, keystroke, and reasoning trace, allowing developers to inspect execution replays and enforce security sandboxing before dispatching production workflows.