Can Radeon AI Pro R9700 run Qwen3.6 27B?
Yes. Measured 66.0 tok/s generation (llama.cpp Vulkan, Q4_K_M + MTP draft head (draft-n-max 3), 64K context; source).
Every measured result for Qwen3.6 27B on Radeon AI Pro R9700
| Build | Quant / setup | Engine | tok/s | |
|---|---|---|---|---|
| Radeon AI Pro R9700: Qwen3.6 27B on Vulkan, HIP and Ollama, with and without MTP | Q4_K_M + MTP draft head (draft-n-max 3), 64K context | llama.cpp Vulkan | 66.0 | source |
| Radeon AI Pro R9700: Qwen3.6 27B on Vulkan, HIP and Ollama, with and without MTP | Q4_K_M, no MTP | llama.cpp Vulkan | 31.9 | source |
| Radeon AI Pro R9700: Qwen3.6 27B on Vulkan, HIP and Ollama, with and without MTP | Q4_K_M | llama.cpp HIP (ROCm) | 27.1 | source |
| Radeon AI Pro R9700: Qwen3.6 27B on Vulkan, HIP and Ollama, with and without MTP | Q4_K_M | Ollama (HIP) | 26.7 | source |
Memory: Radeon AI Pro R9700 has 32 GB. Qwen3.6 27B weights are about 16.7 GB at Q4_K_M (27.32B params × 4.89 bits ÷ 8, before context/KV cache). Arithmetic, not a measurement; long contexts need more. Check other quants in the VRAM calculator.
Speeds are decode (generation) tokens/sec as reported by each source; quant, engine and context differ, so small gaps are noise.
All measured speeds for Qwen3.6 27B: best home rig for Qwen3.6 27B.
Qwen3.6 27B on other hardware: DGX Spark (GB10) · RTX 3090 · RTX 4090 · RTX 5090 · Strix Halo (Radeon 8060S)
Other models measured on Radeon AI Pro R9700: Qwen3.6 35B-A3B · Qwen3.8 27B · gpt-oss 20B
Not enough VRAM? Rent before you buy. Run Qwen3.6 27B on a bigger GPU on a rented 24–96 GB cloud GPU by the hour and see the real speed before spending €1,000+. Try it on Runpod
Referral link: if you sign up with Google and add $10, Runpod gives you and RigBase bonus credit (terms). It never affects our data.
Referral link: if you sign up with Google and add $10, Runpod gives you and RigBase bonus credit (terms). It never affects our data.