v0.15.5-rc0
📦 ollamaView on GitHub →
✨ 3 features
Summary
This release introduces the GLM-OCR model and updates default context sizes based on VRAM availability. It also adds support for GLM-4.7-Flash on the MLX engine.
✨ New Features
- Introduced GLM-OCR, a multimodal OCR model for complex document understanding.
- Added GLM-4.7-Flash support when using Ollama's experimental MLX engine.
- Updated default context size based on available VRAM: 4096 for < 24 GiB, 32768 for 24-48 GiB, and 262144 for >= 48 GiB.