v0.1.0-alpha.1

What's Changed

• Add experimental Qwen2.5 0.5B support by @c8dhjp4tyv-bit in #1
• Add hybrid NPU and CPU layer offload by @c8dhjp4tyv-bit in #2
• Fuse Qwen decoder on XDNA1 by @c8dhjp4tyv-bit in #3

New Contributors

• @c8dhjp4tyv-bit made their first contribution in #1

Full Changelog: https://github.com/c8dhjp4tyv-bit/hawkpoint-npu-llm/commits/v0.1.0-alpha.1