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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |     2048 |  1 |    0 | pp2048 @ d32768 |         14.06 ± 0.00 |
HW Exception by GPU node-1 (Agent handle: 0x97894f0) reason :GPU Hang
✖ ! [rocm-7alpha-rocwmma] GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003 __fa1__longctx32768 failed (exit 134)