Llama 3.2 Vision

Llama 3.2 Vision

by

This experimental pre-release brings initial support for Llama 3.2 Vision. After downloading the Ollama 0.4 pre-release, you can launch it with:

ollama run x/llama3.2-vision

Before the full release of 0.4, we aim to further enhance inference quality and performance and test Ollama’s new Go model runtime. Testing on different platforms and model architectures is welcomed and greatly appreciated.

What’s New

  • Optimized performance on the latest NVIDIA graphics cards (e.g., RTX 40 series)
  • Faster processing for follow-on requests with vision models
  • Fixed issues with correctly detecting stop sequences

Known Issues

  • Sequential image uploads may lead to suboptimal results
  • Errors may occur when running Deepseek models

https://github.com/ollama/ollama/releases/tag/v0.4.0-rc6