verifiedmedium · 0.70updated 2026-10-14

Strix Halo 8060S Vulkan RADV Qwen3.6-35B-A3B UD-Q4_K_XL: pp512 tok/s

1113.64 tok/s (pp512, per-request batch-1)

Fields

benchmarks/ky0-qwen36-35b-a3b-udq4-vulkan-radv-pp512

Metric

prefill_tok_s

Value

1,113.64

Unit

tok/s (pp512, per-request batch-1)

Model

Qwen3.6-35B-A3B

Methodology

THIRD-PARTY MEASURED (kyuz0/amd-strix-halo-toolboxes results.json). Ryzen AI Max+ 395 + Radeon 8060S (gfx1151), 128GB UMA, Fedora 43 kernel 6.19.12. llama.cpp build 9187/9193, env 'vulkan_radv' (Vulkan RADV), -fa 1 -ngl 99 --mmap 0, pp512 + tg128 mean over repeats, per-request batch-1. Model Qwen3.6-35B-A3B (35.51B, 21.27 GiB).

Measured by

third_party

Reproducible

yes

Cross-references

Strix Halo 8060S Vulkan RADV Qwen3.6-35B-A3B UD-Q4_K_XL: pp512 tok/s

Fields

Cross-references

Accelerators (2)

Engine (1)

Interconnect (1)