video-understanding

dsh-video-storyboard

+60

Scene-aware storyboard for DeepSeek Harness: 9-16 numbered cells with timestamps + transcript + an audio envelope, packed into ONE image (384 tokens). Two tools: video_storyboard and video_zoom.

OmniJev/OneJev

+28

🚀🚀 A multimodal System One decision model that gives calibrated answers to typed questions about screens, photos, video and text in one forward pass.

OmniCompass

+26

OmniCompass: A Holistic Benchmark for Omni-Modal Video Grounding and Understanding

lmms-eval

+14

One-for-All Multimodal Evaluation Toolkit Across Text, Image, Video, and Audio Tasks

watch-skill

+16

Give AI agents eyes, ears, and verifiable results. Watch Skill turns video, audio and screen activity into searchable, timestamped evidence and proves work with deterministic contracts, not model opinion. DeepWatch is the agent workspace built on DeepSeek Harness. Python + npm, MCP, CLI, REST, Web.

video-search-and-summarization

+10

NVIDIA AI Blueprint for video search and summarization (VSS) is a GPU-accelerated reference architecture for building video analytics agents with real-time verified alerts, visual Q&A, and automated reporting. The VSS Blueprint uses vision language models (VLMs) such as NVIDIA Cosmos, LLMs such as NVIDIA Nemotron, RAG, and NVIDIA NIMs.

Code-as-World

+12

Code as World: Agentic Discovery of Executable World Representations for Physical Reasoning

SALMONN

+9

SALMONN family: A suite of advanced multi-modal LLMs

zai-org/GLM-V

+3

GLM-4.6V/4.5V/4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning