# PKU-YuanGroup/Video-LLaVA

【EMNLP 2024🔥】Video-LLaVA: Learning United Visual Representation by Alignment Before Projection

Repository: https://github.com/PKU-YuanGroup/Video-LLaVA
Canonical: https://ross.abutalabs.com/products/video-llava
Homepage: https://arxiv.org/pdf/2311.10122.pdf
Language: Python
License: Apache-2.0
License Family: permissive
Topics: instruction-tuning, large-vision-language-model, multi-modal
Last push: 2024-12-03T02:58:46+00:00

## Health v2 (maintenance only)
Score: 27/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 74
- inputs: {"age_days": 1045, "days_push": 638, "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 3500, forks 255 (observed 2026-08-28T04:08:07.494120+00:00)

## What it is
Video-LLaVA is a large vision-language model that aligns image and video representations into a unified visual space before projection into the language model, enabling joint image and video understanding. It is an EMNLP 2024 research codebase with pretrained checkpoints, training scripts, and demo spaces.

## Use cases
- answer questions about videos with an llm
- video question answering model
- chat with images and videos using a vision-language model
- run multimodal llm inference on video
- fine-tune a video-language model with instruction tuning
- unified image and video understanding model

## When to choose
- you need a single model that handles both image and video inputs
- you want an open-source research baseline for video-language alignment
- you need video question answering or captioning with an LLM

## When to avoid
- you need production-grade, commercially supported multimodal APIs
- you lack a GPU for inference or training
- you only need text-only LLM capabilities

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, llm-inference, video-processing, image-processing, nlp
- domain: large-language-models, computer-vision, artificial-intelligence
- platform: python
- tags: multimodal, vision-language-model, video-understanding, instruction-tuning, video-qa, research, video, natural-language-processing, gpu, linux

## Member repositories
- PKU-YuanGroup/Video-LLaVA (main) score 27

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:07.494120+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 2026-08-29T18:35:48.427696+00:00, confidence not recorded.
  - readme: https://github.com/PKU-YuanGroup/Video-LLaVA (fetched 2026-08-28T04:08:07.494120+00:00, sha 6c3ae86af9ee)
- Data as of 2026-08-30T08:39:29.467469+00:00.