GLM-4.6V / b8118 Vulkan / PP512 TG128 at prepared depth8000
AMD Radeon 8060S gfx1151 (Strix Halo; CPU SKU and installed RAM unknown) · llama.cpp
- generation:
- 11.1 tokens/s
- prompt processing (prefill):
- 118.0 tokens/s
- quant:
- UD-Q4_K_XL (GGUF)
- kv:
- Q8
- flash attention:
- on
- mtp (multi-token prediction):
- off
Reported by the source; GPU count, offloading and concurrent requests can change these figures. Prompt-processing speed is input throughput, not time to first token. Check the full setup before comparing.
User reports GLM-4.6V at 11.1 t/s generation and 118 t/s prompt processing on a single Radeon 8060S gfx1151 device. Setup is llama.cpp b8118 with UD-Q4_K_XL, q8_0 KV cache, flash attention, and ngl 999; the headline figures come from the Vulkan/RADV, BLAS arm at 118.01/11.11 tok/s. Measurements are synthetic llama-bench runs at depth 8000, with PP512 and TG128 as separate tests; no image input or vision projector execution was measured.