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 | pp512 | 296.33 ± 1.49 | | llama4 17Bx16E (Scout) Q8_0 | 106.65 GiB | 107.77 B | ROCm | 99 | 2048 | 1 | 0 | tg128 | 12.17 ± 0.00 | build: ee8dd5c65 (7035)