GetStream/Vision-Agents
Open Vision Agents by Stream is a platform for building real-time, low-latency multi-modal AI agents that understand and interact with video content using advanced models and Stream's edge network.
Awesome AI Agents › Video Processing Agents
HKUDS/VideoAgent
VideoAgent from HKUDS is a research framework that treats video understanding, editing and generation as a single agentic pipeline driven by conversation. Its stated goal is that a user describes what they want in natural language and the system produces the corresponding video artifact without a specialized editing interface. Three components are described. Intent analysis decomposes an instruction into explicit and implicit sub-intents, capturing requirements the user did not state directly, and an intent-to-agent mapping step selects which capabilities of the multi-agent framework need to be activated so that unnecessary components are not run. Autonomous tool use and planning translates the resulting intents into executable workflows through a graph, where nodes represent tool capabilities and edges define the connections between them; the system chooses agents and orders their execution dynamically, and adaptive feedback loops refine the plan through a two-step self-evaluation. Multi-modal understanding converts raw input into semantically aligned visual queries to improve retrieval over the source material. The functional surface covers question answering over video and summarization, editing operations such as assembling and clipping footage, beat-synced and commentary edits, and generation of new material including meme videos, music videos, song remixes, storytelling clips and cross-lingual or cross-cultural adaptations, along with sound effect tooling. A comparison table in the README positions the project against Director, Funclip, NarratoAI and NotebookLM, claiming coverage of editing and remaking capabilities that those tools do not provide. The repository is accompanied by an arXiv paper, demonstration videos, an evaluation section and quick start instructions. It is aimed at researchers and practitioners working on multi-modal agents and automated video production rather than at end users looking for a finished editing application.
https://github.com/HKUDS/VideoAgent
Also in Video Processing Agents
RayVentura/ShortGPT
ShortGPT is an AI-powered framework that automates video content creation and editing for platforms like YouTube Shorts and TikTok, supporting multi-language voiceovers and automated footage sourcing.
showlab/Awesome-Video-Diffusion
A curated list of recent diffusion models and resources for video generation, editing, restoration, and various AI-driven video applications.
dramaclaw/dramaclaw
Source-available AIGC video production pipeline that turns a manuscript into a finished short drama, handling character extraction, episode planning, scripts, storyboards, voice-over and final editing.
nexu-io/html-video
Apache-2.0 toolkit that lets local coding agents turn HTML, CSS and data into real MP4 video through pluggable render engines, 21 animated templates and an optional AI soundtrack.
video-db/Director
Director is an AI video agents framework that enables complex video tasks like search, editing, and generation through natural language commands and real-time streaming, built on VideoDB's video-as-data infrastructure.
video-db/StreamRAG
StreamRAG is a GPT-powered video search and streaming agent that enables real-time video search, summarization, and publishing of searchable video collections on the ChatGPT store.
microsoft/DeepVideoDiscovery
Microsoft research agent for long-form video question answering that treats segmented clips as an environment and iteratively selects multi-granular search tools, reaching state of the art on LVBench.