# mbzuai-oryx/Video-ChatGPT

[ACL 2024 🔥] Video-ChatGPT is a video conversation model capable of generating meaningful conversation about videos. It combines the capabilities of LLMs with a pretrained visual encoder adapted for spatiotemporal video representation. We also introduce a rigorous 'Quantitative Evaluation Benchmarking' for video-based conversational models.

Repository: https://github.com/mbzuai-oryx/Video-ChatGPT
Canonical: https://ross.abutalabs.com/products/video-chatgpt
Homepage: https://mbzuai-oryx.github.io/Video-ChatGPT
Language: Python
License: CC-BY-4.0
License Family: other
Topics: chatbot, clip, gpt-4, llama, llava, mulit-modal, vicuna, vision-language, vision-language-pretraining, video-chatboat, video-conversation
Last push: 2025-08-05T01:07:25+00:00

## Health v2 (maintenance only)
Score: 45/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 35, release rhythm 35, longevity 85
- inputs: {"age_days": 1203, "days_push": 394, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1506, forks 128 (observed 2026-08-28T04:04:55.106343+00:00)

## What it is
Video-ChatGPT is a video conversation model that combines large language models with a pretrained visual encoder adapted for spatiotemporal video representation, enabling detailed conversation about videos. It also ships a quantitative evaluation benchmarking framework (VCGBench) for assessing video-based conversational models.

## Use cases
- chat with a model about a video's content
- generate detailed video descriptions and captions
- zero-shot video question answering
- benchmark video conversational models quantitatively
- build a multimodal video understanding assistant
- evaluate temporal understanding of video LLMs

## When to choose
- you need a research-grade video-language conversation model
- you want a standardized benchmark for video chat models
- you need zero-shot video QA on datasets like MSVD, MSRVTT, TGIF, or ActivityNet

## When to avoid
- you need production-ready, low-latency video analysis at scale
- you lack GPU resources for inference and training
- you only need image (not video) understanding

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, video-processing, chatbot, llm-inference, benchmarking
- domain: artificial-intelligence, computer-vision, large-language-models
- platform: python
- tags: multimodal, vision-language-model, video-understanding, video-qa, video-conversation, evaluation-benchmark, research, video, natural-language-processing, gpu, linux

## Member repositories
- mbzuai-oryx/Video-ChatGPT (main) score 45

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:55.106343+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:32:40.142188+00:00, confidence not recorded.
  - readme: https://github.com/mbzuai-oryx/Video-ChatGPT (fetched 2026-08-28T04:04:55.106343+00:00, sha f8dd666e4b74)
  - homepage: https://mbzuai-oryx.github.io/Video-ChatGPT (fetched 2026-08-29T11:37:13.652181+00:00, sha 1f7658b62290)
- Data as of 2026-08-30T08:39:29.467469+00:00.