# AMD GPU Support

**URL:** https://discuss.vllm.ai/c/hardware-support/vllm-amd-gpu-support/7.md

[Latest](https://discuss.vllm.ai/latest.md) · [Categories](https://discuss.vllm.ai/categories.md)

---

## [About the AMD GPU Support category](https://discuss.vllm.ai/t/about-the-amd-gpu-support-category/64)

<div class="topic-metadata">

**Author:** [@youkaichao](https://discuss.vllm.ai/u/youkaichao)\
**Replies:** 0

</div>

---

## [GFX1030 Support PR](https://discuss.vllm.ai/t/gfx1030-support-pr/2930)

<div class="topic-metadata">

**Author:** [@BlivionIaG](https://discuss.vllm.ai/u/BlivionIaG)\
**Replies:** 1\
**Last updated:** [August 27, 2026, 10:07pm UTC](https://discuss.vllm.ai/t/gfx1030-support-pr/2930 "2026-08-27T22:07:49Z")

</div>

Hello there, I couldn’t find a space or place to ask about where to discuss the gfx1030 pr feat(rocm): enable gfx1030 (RDNA2) as a recognized RDNA arch by BlivionIaG · Pull Request #52391 · vllm-project/vllm · GitHub R…

---

## [Win11 wsl2 在 AMD RYZEN AI MAX+ 395](https://discuss.vllm.ai/t/win11-wsl2-amd-ryzen-ai-max-395/2752)

<div class="topic-metadata">

**Author:** [@kuolung1](https://discuss.vllm.ai/u/kuolung1)\
**Replies:** 1\
**Last updated:** [July 3, 2026, 1:10am UTC](https://discuss.vllm.ai/t/win11-wsl2-amd-ryzen-ai-max-395/2752 "2026-07-03T01:10:23Z")

</div>

在以上的環境中， 無法正常跑 DeepSeek ocr 2 我想掛 ROCm 使用8060s GPU

---

## [Why does quickReduce not need to use system-scope release write operations to update flags?](https://discuss.vllm.ai/t/why-does-quickreduce-not-need-to-use-system-scope-release-write-operations-to-update-flags/1370)

<div class="topic-metadata">

**Author:** [@sandstorm666](https://discuss.vllm.ai/u/sandstorm666)\
**Replies:** 0\
**Last updated:** [August 13, 2025, 4:22pm UTC](https://discuss.vllm.ai/t/why-does-quickreduce-not-need-to-use-system-scope-release-write-operations-to-update-flags/1370 "2025-08-13T16:22:25Z")

</div>

\_quickreduce\_device\_inline\_ void set\_sync\_flag(uint32\_t\* flag\_ptr, uint32\_t flag) { \_\_atomic\_store\_n(flag\_ptr, flag, \_\_ATOMIC\_RELEASE); } \_quickreduce\_device\_inline\_ void …

---

## [How to setup amd gpu as default in dual stack gpu?](https://discuss.vllm.ai/t/how-to-setup-amd-gpu-as-default-in-dual-stack-gpu/383)

<div class="topic-metadata">

**Author:** [@CarlosR759](https://discuss.vllm.ai/u/CarlosR759)\
**Replies:** 10\
**Last updated:** [April 21, 2025, 7:08pm UTC](https://discuss.vllm.ai/t/how-to-setup-amd-gpu-as-default-in-dual-stack-gpu/383 "2025-04-21T19:08:40Z")

</div>

Hi I’m having issues with vllm because I have a two stack gpu in a machine. One is nvidia and another is amd. My nvidia card is an older one but maybe I’m going to use it as an accelerator for speculative decoding with …