# DAMO-NLP-SG/Video-LLaMA

[EMNLP 2023 Demo] Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding

Repository: https://github.com/DAMO-NLP-SG/Video-LLaMA
Canonical: https://ross.abutalabs.com/products/video-llama
Language: Python
License: BSD-3-Clause
License Family: permissive
Topics: large-language-models, video-language-pretraining, vision-language-pretraining, blip2, llama, minigpt4, cross-modal-pretraining, multi-modal-chatgpt
Last push: 2024-06-04T07:06:41+00:00

## Health v2 (maintenance only)
Score: 29/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 86
- inputs: {"age_days": 1215, "days_push": 820, "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 3139, forks 287 (observed 2026-08-28T04:07:45.796690+00:00)

## What it is
Video-LLaMA is an instruction-tuned audio-visual language model that extends LLaMA with video and audio understanding via cross-modal pretraining (BLIP-2 style Q-former). It provides pretrained and finetuned checkpoints plus inference code for multimodal video chat.

## Use cases
- chat with a video using an LLM
- understand video content with a multimodal model
- ask questions about a video's audio and visuals
- build a video question-answering demo
- finetune a video-language model on custom data
- summarize videos with a large language model

## When to choose
- you need open checkpoints for video+audio instruction-following chat
- you want a research baseline for video-language understanding
- you want to customize a pretrained video LLM with your own data

## When to avoid
- you need the strongest or easiest-to-use codebase - the authors recommend VideoLLaMA2
- you need production-grade, actively maintained software
- you lack GPU resources for 7B/13B multimodal inference
- you need Chinese-language chat support

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, llm-inference, video-processing, audio-processing, chatbot
- domain: large-language-models, computer-vision, artificial-intelligence
- platform: python
- tags: multimodal, video-understanding, audio-visual, instruction-tuning, llama, blip2, vision-language-model, research-model, video, natural-language-processing, gpu, linux

## Member repositories
- DAMO-NLP-SG/Video-LLaMA (main) score 29

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:45.796690+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 2026-08-30T07:26:02.867502+00:00, confidence not recorded.
  - readme: https://github.com/DAMO-NLP-SG/Video-LLaMA (fetched 2026-08-28T04:07:45.796690+00:00, sha de85c22c827a)
- Data as of 2026-08-30T08:39:29.467469+00:00.