AMD Adds Gemma 4 Support Across Hardware
AMD on April 4, 2026 announces Day One support for Google's Gemma 4 models across its Radeon GPUs, Instinct datacenter GPUs, and Ryzen AI CPUs. AMD details integrations with vLLM, SGLang, LM Studio, llama.cpp, Lemonade and ROCm, and provides deployment guidance for MI300-series GPUs and XDNA 2 NPU. The move enables immediate local and datacenter inference for Gemma 4 models (2B–31B).
Key Points
- 1Announces Day Zero support for Gemma 4 across Radeon, Instinct, and Ryzen AI hardware.
- 2Enables optimized inference paths via vLLM, SGLang, ROCm, Lemonade, and LM Studio integrations.
- 3Allows developers to run Gemma 4 models locally and at scale on AMD GPUs and NPUs.
Scoring Rationale
Official Day Zero support from AMD provides broad, actionable deployment paths and strong credibility, boosting scope and actionability. Novelty is limited because Gemma 4 is Google's launch; score reflects high operational impact and official integrations with major frameworks.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
