●Change8

v0.15.5-rc1

📦 ollamaView on GitHub →
✨ 2 features🔧 2 symbols

Summary

This release introduces the GLM-OCR model and adds support for GLM-4.7-Flash on the MLX engine, while also updating default context sizes based on VRAM.

Migration Steps

  1. Review default context size settings as they have been updated based on available VRAM.
  2. If using the experimental MLX engine, GLM-4.7-Flash is now available.

✨ New Features

  • Introduced GLM-OCR, a multimodal OCR model for complex document understanding.
  • Added support for GLM-4.7-Flash on Ollama's experimental MLX engine.

Affected Symbols

GLM-OCRGLM-4.7-Flash