AMD Rolls Out Gemma 4 Model Support Across Full Range of GPUs & CPUs

AMD announced Day Zero support for Google’s Gemma 4 across its Radeon GPUs, Instinct GPUs, and Ryzen AI CPUs.

The company said Gemma 4 integrates with tools like vLLM, SGLang, LM Studio, llama.cpp, Ollama, and Lemonade.

vLLM and SGLang provide Docker and Python deployment paths and require the Triton attention backend for certain attention features.

Gemma 4 can fit on a single MI300X GPU at TP=1 with full context, and NPU support for E2B/E4B will arrive in a forthcoming Ryzen AI software update.