Expired60% confidenceFactExact time
AI inference support for AMD graphics cards relies on the ROCm platform, which mainly supports RDNA2/RDNA3 architecture cards from the RX 6000 series and above; older cards may not receive official support.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
7/22/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/20/2026(expired)
Sources
Ollama本地部署大模型：Windows完整安装教程
Related Claims
UnverifiedGPU models supported by ROCm primarily cover the Radeon RX 6000/7000 series and Instinct data center cards, while some consumer-grade graphics cards such as the RX 5000 series are at the edge of or absent from the official support list.74% similarUnverifiedOllama and llama.cpp support for AMD GPUs relies on the ROCm driver stack, and ROCm compatibility for integrated GPUs on laptop platforms is not yet fully mature.73% similarUnverifiedAMD's MI300X surpasses the H100 in memory bandwidth and provides a degree of CUDA compatibility through the ROCm platform.73% similarUnverifiedAMD Instinct系列（如MI300X）通过ROCm软件栈实现与PyTorch、TensorFlow等主流AI框架的兼容73% similarUnverifiedThe 7900XTX uses the RDNA 3 architecture and has received official support in ROCm version 6.x.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/590881API
curl https://kongchang.com/api/v1/knowledge/claims/590881MCP
get_claim(id=590881)