v0.32.3

playcanvas/enginev0.32.3Jul 23, 2026by github-actions[bot]

AI Summary

This update improves stability by fixing model download stalling and restores integrations for Claude Code Channels and Anthropic thinking streams. It expands hardware support to Windows ARM64 and B200 GPUs while adding new capabilities for Laguna 2.1 models.

Key Highlights

Fixed model downloads that stall before sending data
Expanded GPU support including Windows ARM64 and B200 (CUDA 12)
Added chat, thinking, and tool calling support for Laguna 2.1
Improved Anthropic thinking streams and Claude Code Channels

New Features

Chat, thinking, and tool calling for Laguna 2.1
Windows ARM64 CUDA support
B200 GPU support

Full Release Notes

## What's Changed

- Fixed model downloads that stall before sending data.
- Improved integrations: restored Claude Code Channels, fixed Anthropic thinking streams, and made Hermes Desktop respect `--force-build`.
- Expanded GPU support with CUDA on Windows ARM64, B200 support through CUDA 12, and lower memory use on Linux CUDA/ROCm iGPUs.
- Added chat, thinking, and tool calling support for Laguna 2.1 models, including a Metal inference fix.
- Fixed GLM tool calls being silently dropped at the end of generation.
- Updated the MLX and llama.cpp engines.

**Full Changelog**: https://github.com/ollama/ollama/compare/v0.32.1...v0.32.3