Quick start
ollama run llama3.2-visionAvailable sizes
| Tag | Size | Quantization | Context | Min RAM |
|---|---|---|---|---|
| llama3.2-vision:latest | 7.8GB | q4_k_m | 128K context | 9.8 GB |
| llama3.2-vision:90b | 55GB | q4_k_m | 128K context | 68.8 GB |
Strengths & Limitations
Strengths
- Image reasoning
- Generative modeling
- Instruction following
Related models
gemma3General
The current, most capable model that runs on a single GPU.
32.1M pullsllavaMultimodal
🌋 LLaVA is a novel end-to-end trained large multimodal model that combines a vision encoder and Vicuna for general-purpose visual and language understanding. Updated to version 1.6.
12.9M pullsminicpm-vMultimodal
A series of multimodal LLMs (MLLMs) designed for vision-language understanding.
4.6M pullsllava-llama3Multimodal
A LLaVA model fine-tuned from Llama 3 Instruct with better scores in several benchmarks.
2.1M pulls