Builds / AMD Ryzen AI Max+ 395 (Radeon 8060S iGPU)
Strix Halo (Corsair AI Workstation 300, 128 GB): ROCm vs Vulkan on Qwen3.8 27B and Qwen3.6 35B-A3B
Source: howquant.dev ↗📏 How measured: reported by the original source; RigBase re-checked the numbers at that source
- GPU(s)
- AMD Ryzen AI Max+ 395 (Radeon 8060S iGPU)
Amazon · eBay (used) · Newegg - VRAM / unified memory (GB)
- 128
- CPU
- AMD Ryzen AI Max+ 395
Amazon · eBay (used) · Newegg - Motherboard
- Corsair AI Workstation 300
Amazon · eBay (used) · Newegg - RAM
- 128 GiB (8 x 16 GB, 'LPDDR5' per dmidecode, 8000 MT/s configured)
Amazon · eBay (used) · Newegg - PSU
- not stated
- Other
- Ubuntu 24.04, ROCm 7.14, Mesa RADV 26.1.7, llama.cpp c060ca974
- Reported total cost
- not stated
- Power
- Balanced power profile
Measured speed (generation tokens/sec)
| Model | Quant | Engine | tok/s | Details |
|---|---|---|---|---|
| Qwen3.6 35B-A3B | Q8_0, llama-bench tg128 | llama.cpp Vulkan | 54.22 | |
| Qwen3.6 35B-A3B | Q8_0, llama-bench tg128 | llama.cpp ROCm | 47.47 | |
| Qwen3.6 35B-A3B | Q8_0 + MTP (draft-n-max 3), coding workload | llama.cpp ROCm | 60.92 | ⚡ spec-decode (MTP) |
| Qwen3.8 27B | Q8_0, llama-bench tg128 | llama.cpp ROCm | 7.76 | |
| Qwen3.8 27B | Q8_0, llama-bench tg128 | llama.cpp Vulkan | 7.77 | |
| Qwen3.8 27B | Q8_0 + MTP (draft-n-max 3), coding workload, reasoning on | llama.cpp ROCm | 15.67 | ⚡ spec-decode (MTP) |
| Qwen3.8 27B | Q8_0 + MTP (draft-n-max 3), coding workload, reasoning on | llama.cpp Vulkan | 13.19 | ⚡ spec-decode (MTP) |
How Quant, first published 24 Aug 2026, updated 16 Sep 2026. llama-bench means over 5 runs; MTP rows are llama-server runs on SPEED-Bench's coding workload (Vulkan MoE + MTP measured 60.40). Same llama.cpp commit for both backends.
Source: https://howquant.dev/benchmarks/strix-halo-rocm-vs-vulkan/
Blank fields = not stated in the source. We never estimate build data; GPU street prices above are separately sourced and dated.
Rent before you buy: try this model on a cloud GPU for a few dollars first: Runpod (referral link; sign up with Google to get the new-user bonus)
Quick answers: Can Strix Halo (Radeon 8060S) run Qwen3.6 35B-A3B? · Can Strix Halo (Radeon 8060S) run Qwen3.8 27B?