ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 ROCm devices:
  Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model                          |       size |     params | backend    | ngl | n_ubatch | fa | mmap |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| llama4 17Bx16E (Scout) Q8_0    | 106.65 GiB |   107.77 B | ROCm       |  99 |     2048 |  1 |    0 | pp2048 @ d32768 |        190.45 ± 0.00 |
HW Exception by GPU node-1 (Agent handle: 0x145324f0) reason :GPU Hang
✖ ! [rocm-7alpha] Llama-4-Scout-17B-16E-Instruct-Q8_0-00001-of-00003 __fa1__longctx32768 failed (exit 134)