vLLM on the AMD Radeon AI PRO R9700 32GB
On the AMD Radeon AI PRO R9700 32GB, vLLM runs Qwen3.6 35B · 3B active at 4-bit at a median 61.4 tokens per second for generation, from 1 plain run (256K context).
Speed by model on the AMD Radeon AI PRO R9700 32GB
Plain runs only: one device, one request, no speculative decoding, the whole model in memory. Each row is one model size at one bit class, with the other engines' plain runs of the same model on this card beside it. Context length and build still differ between runs, and each side shows its own.
| Model | vLLM | Other engines |
|---|---|---|
| Qwen3.6 35B · 3B active 4-bit | 61.4 t/sone report, 256K context | no other engine reported |
13 vLLM reports on the AMD Radeon AI PRO R9700 32GB
Newest first. A run on several cards, with speculative decoding, with part of the model in system RAM or with batched requests says so, since each describes a different setup.
- Qwen3.8 125B · 6B active · MXFP4-FP8-GPTQ70.0 t/s(4 cards, speculative decoding)
- Qwen3.8 27B · W3A4179.7 t/s(speculative decoding)
- Qwen3.8 27B · MXFP4113.0 t/s(2 cards)
- Qwen3.8 27B · MXFP4184.9 t/s(2 cards)
- Qwen3.8 27B · NVFP4276.0 t/s
- Qwen3.8 27B · NVFP4120.0 t/s
- Qwen3.8 27B · Q890.8 t/s
- Qwen3.6 35B · 3B active · INT461.4 t/s
- Qwen3.8 125B · 6B active · MXFP4-FP8120.0 t/s(4 cards)
- Qwen3.8 27B · FP856.6 t/s(speculative decoding)
- Qwen3.8 27B · FP875.2 t/s(2 cards)
- Qwen3.8 27B · MXFP4280.0 t/s(2 cards, speculative decoding)
- Qwen3.8 27B · FP887.6 t/s(2 cards, speculative decoding)
Frequently asked
How fast is vLLM on the AMD Radeon AI PRO R9700 32GB?
On the AMD Radeon AI PRO R9700 32GB, vLLM runs Qwen3.6 35B · 3B active at 4-bit at a median 61.4 tokens per second for generation, from 1 plain run (256K context).