LLM EXPLORER 65,894 MODELS INDEXED

LLaVA OneVision 2 8B Instruct by lmms-lab-encoder

By lmms-lab-encoder · 10532 downloads
HF Repo ↗ Copy URL

LLaVA OneVision 2 8B Instruct is an open-source language model by lmms-lab-encoder. Features: 8b LLM, VRAM: 17GB, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.29.

  Arxiv:2509.23661   Arxiv:2602.08683   Arxiv:2605.25979   Conversational   Custom code   En   Image-text-to-text   Instruct   Llava   Llava-onevision-2   Llava onevision2   Multimodal   Qwen3   Region:us   Safetensors   Sharded   Tensorflow   Video-text-to-text   Vision   Vision-language   Zh
Model Card on HF 🤗: https://huggingface.co/lmms-lab-encoder/LLaVA-OneVision-2-8B-Instruct 

LLaVA OneVision 2 8B Instruct Parameters and Internals

LLM NameLLaVA OneVision 2 8B Instruct
Repository 🤗https://huggingface.co/lmms-lab-encoder/LLaVA-OneVision-2-8B-Instruct 
Model Size8b
Required VRAM17 GB
Updated2026-09-12
Maintainerlmms-lab-encoder
Model Typellava_onevision2
Instruction-BasedYes
Model Files  5.3 GB: 1-of-4   4.5 GB: 2-of-4   5.3 GB: 3-of-4   1.9 GB: 4-of-4
Supported Languagesen zh
Model ArchitectureLlavaOnevision2ForConditionalGeneration
Licenseapache-2.0
Model Max Length131072
Transformers Version5.7.0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace