DeerFlow is an open-source, full-stack AI agent harness built on LangGraph/LangChain, providing a production-ready platform for building, deploying, and managing conversational AI agents. The system features a Python/FastAPI backend with a modular architecture that includes agent orchestration, long-term memory (DeerMem), sandboxed code execution, a skill system with static analysis (SkillScan), MCP (Model Context Protocol) tool integration, multi-channel IM support (Telegram, Slack, Discord, Lark, etc.), scheduled tasks, subagent batches, and a comprehensive authorization framework. The frontend is a Next.js/React workspace with real-time streaming, artifact management, and a terminal UI (TUI) for headless operation. The architecture follows a gateway pattern: the FastAPI gateway handles authentication (JWT, OAuth, PATs), routing, and SSE streaming, while the core harness package (`deerflow`) implements agent logic, middleware chains, and runtime components. Persistence uses SQLAlchemy with Alembic migrations, supporting both SQLite and PostgreSQL. The system is highly extensible via a Python extension API (`deerflow-extension-api`) and a plugin system for frontend surfaces. Deployment supports Docker Compose and Kubernetes (Helm chart). While primarily an application, DeerFlow includes research-oriented components such as memory eviction benchmarks, context snapshot evaluations, and tracing integrations (LangSmith, Langfuse, OpenTelemetry). The codebase demonstrates strong engineering practices with extensive unit tests, blocking IO regression tests, and contract tests, though its large feature surface and many configuration options introduce complexity and potential for drift between environments.

This repository is a large monorepo for a multimodal AI agent platform developed by ByteDance. It includes a desktop Electron application (UI-TARS Desktop) that allows users to control their computer using natural language via vision-language models (UI-TARS, GPT-4o, Claude, etc.). Additionally, it provides a CLI tool (Agent TARS) for running agents in browser/desktop environments, a server for multi-user agent sessions, a web UI for interacting with agents, MCP servers for browser/filesystem/commands, and a comprehensive SDK for building custom agents and operators. The architecture is heavily componentized with multiple packages for agent core, tool-calling engines (native, prompt-engineering, structured-outputs), browser automation (local, remote, sandbox), desktop automation (NutJS, ADB), and UI components. While the codebase demonstrates strong engineering with clean abstractions (event streams, plugin systems, strategy patterns), it also shows signs of rapid growth with some duplication (two agent server implementations, overlapping operator packages) and ongoing migrations. The repository is actively maintained with frequent releases and comprehensive test coverage across unit, integration, and e2e tests.

xgplayer is a production-grade, plugin-based HTML5 video player library maintained by ByteDance. It provides a core player package (`packages/xgplayer`) with UI controls, event/hook systems, device sniffing, i18n, and a rich plugin registry, plus separate npm packages for streaming protocols including HLS, FLV, DASH, MP4, and MPEG-TS transmuxing. The architecture is a monorepo of coordinated packages: streaming plugins built on shared networking, MSE, buffer, and error utilities; a custom type-level transmuxer that demuxes/remuxes FLV/TS/MP4 into fMP4 for Media Source Extensions; and extension packages for Chromecast/AirPlay, Google IMA ads, subtitles, and music visualization. In production this library plays both VOD and live streams in the browser, handles adaptive bitrate switching, DRM, low-latency manifest polling, and provides full control UI for desktop and mobile.

Dolphin is a document image parsing tool that uses a two-stage analyze-then-parse approach with a VLM (Qwen2.5-VL). It takes images or PDFs, performs layout analysis to detect elements (text, tables, formulas, code, figures) with reading order, then parses each element in parallel batches. Outputs structured JSON and Markdown with visualized layout. The codebase is well-structured, offering CLI entry points for page-level, element-level, and layout-only parsing. It demonstrates good engineering practices with batching, coordinate mapping, and post-processing for formula/special characters. The core implementation wraps the Hugging Face Qwen2.5-VL model and includes utility modules for layout string parsing, PDF conversion, and markdown generation. The architecture is a CLI application with minimal coupling, making the utility modules potentially reusable across document parsing projects.

FlowGram is a comprehensive, production-grade visual workflow development framework from ByteDance. The codebase implements two distinct editing paradigms — fixed-layout (automatic positioning with composite nodes like loops, branches) and free-layout (node-position-free with lines) — both built on a shared plugin architecture using inversify for dependency injection. The framework includes a form engine with reactive fields, validation, and effects; a variable engine with AST-based type inference and scoped lifecycle; and a runtime engine capable of executing workflows with executors for LLM, HTTP, code (QuickJS sandbox), and conditions. The repository contains over 40 packages managed via Rush monorepo, encompassing core libraries, plugins (history, minimap, auto-layout, group, snap, clipboard, export), UI material libraries (Semi UI, Ant Design), a CLI tool for scaffolding, multiple demo applications, and comprehensive documentation. The system is actively maintained with regular commits, CI workflows, and E2E tests. Notable technical decisions include a local fork of the use-gesture library for canvas interaction, heavy decorator-based IoC, and support for both React 17 and 18. While the architecture is well-structured and extensible, the codebase shows signs of duplication between fixed and free layouts, numerous demo-specific components, and some deprecated APIs that indicate ongoing refactoring.

Protenix is a production-oriented PyTorch implementation of AlphaFold 3 for biomolecular structure prediction, maintained by ByteDance. It is best classified as an application rather than a pure research codebase: it ships a `protenix` CLI with commands for prediction, input conversion, MSA/template/RNA-MSA search, and data preparation, plus training/fine-tuning entry points, batch inference runners, Docker deployment, web-service helpers, and PyPI packaging. The code ingests PDB/CIF, FASTA, SMILES/SDF, and JSON descriptors covering proteins, DNA, RNA, ligands, ions, modified residues, covalent bonds, and contact/pocket constraints, then outputs predicted 3D CIF structures with confidence metrics such as pLDDT, PAE, pTM, and ipTM.

Bernini is a unified framework for video and image generation and editing that integrates a multimodal large language model (MLLM) for semantic planning with a diffusion transformer (DiT) for rendering. It supports tasks including text-to-image, image-to-image, text-to-video, video-to-video, reference-to-video, and reference-video-to-video. The architecture features modular pipelines for both full (with MLLM) and renderer-only modes, leveraging the VeOmni framework for distributed training and inference via Ulysses sequence parallelism. Key components include custom attention backend selection, flow matching scheduler, and utilities for weight loading and data processing.