Tag
The NPU on AMD Strix Halo devices is now usable for AI inference, enabling hybrid mode that combines NPU and iGPU for faster prompt processing. Tools like Lemonade and AMD's ROCm software make this possible.
AMD's Lemonade SDK for local AI adds NVIDIA CUDA support in version 10.7, enabling the same local AI server experience on competitor GPUs. The release also introduces lemonade bench for cross-backend LLM benchmarking and broader Vulkan support.
Lemonade v10.5.1 adds MTP support and ROCm 7.13 quick start for Strix Halo, along with a Fedora 43 fix.
Lemonade has added an experimental ROCm backend for vLLM, allowing users to easily run safetensors LLMs on AMD GPUs with a simple command.
Community maintainer integrates Lemonade local LLMs with Claude Code and other CLIs, enabling local LLM usage.