Llama 3.2 Vision

This pre-release adds initial experimental support for Llama 3.2 Vision. After downloading the Ollama 0.4 pre-release, run:

ollama run x/llama3.2-vision

Before fully releasing 0.4, our goal is to improve inference quality and performance, alongside testing Ollama's new Go model runtime (see below). Any help testing this new release across different platforms and model architectures would be greatly appreciated.

New Go runtime

0.4 is also built on Ollama's new Go runtime, which includes improved vision model caching, model reliability, caching and stop token detection among other fixes. Sending follow-on requests to image models should be much faster.

Notes & known issues

• Sending follow on images can cause issues in response quality
• 0.4 file sizes are larger to better support a wider array of NVIDIA graphics cards with faster performance

Full Changelog: https://github.com/ollama/ollama/compare/v0.3.14...v0.4.0-rc3