Updated benchmarks

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm6_4_3-rocwmma.log

@@ -9,7 +9,7 @@ rocBLAS warning: hipBlasLT failed, falling back to tensile.
 This message will be only be displayed once, unless the ROCBLAS_VERBOSE_TENSILE_ERROR environment variable is set.
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        126.62 ± 0.10 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        125.93 ± 0.26 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         19.95 ± 0.02 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.52 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm6_4_3-rocwmma__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        135.10 ± 0.35 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        135.40 ± 0.23 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.14 ± 0.01 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.69 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm6_4_3-rocwmma__hblt0.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        130.99 ± 0.36 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        132.28 ± 0.14 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.14 ± 0.01 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.50 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm6_4_3-rocwmma__hblt0__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        140.15 ± 0.41 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        139.86 ± 0.32 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.15 ± 0.01 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.70 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm6_4_3.log

@@ -9,7 +9,7 @@ rocBLAS warning: hipBlasLT failed, falling back to tensile.
 This message will be only be displayed once, unless the ROCBLAS_VERBOSE_TENSILE_ERROR environment variable is set.
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        126.66 ± 0.22 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        125.92 ± 0.27 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.14 ± 0.00 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.52 ± 0.01 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm6_4_3__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        100.20 ± 0.13 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        134.12 ± 0.59 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.30 ± 0.01 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.66 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm6_4_3__hblt0.log

@@ -2,5 +2,9 @@ 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
-HW Exception by GPU node-1 (Agent handle: 0x2624d340) reason :GPU Hang
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
-✖ ! [rocm6_4_3] GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002 __hblt0 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        131.45 ± 0.35 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.53 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm6_4_3__hblt0__fa1.log

@@ -2,5 +2,9 @@ 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
-Memory access fault by GPU node-1 (Agent handle: 0x37c5d340) on address 0x7f2e3516f000. Reason: Page not present or supervisor privilege.
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
-✖ ! [rocm6_4_3] GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002 __hblt0__fa1 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        140.67 ± 0.26 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.67 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |         94.56 ± 0.11 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         19.90 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants__fa1.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        127.25 ± 0.57 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.66 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants__hblt0.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        128.69 ± 0.57 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.56 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants__hblt0__fa1.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        169.19 ± 0.12 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.67 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc-rocwmma.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        117.48 ± 0.53 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |         94.71 ± 0.12 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.11 ± 0.00 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.53 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc-rocwmma__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        126.27 ± 0.47 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        126.97 ± 0.54 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         19.86 ± 0.00 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.70 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc-rocwmma__hblt0.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        158.54 ± 0.42 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        160.39 ± 0.34 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.11 ± 0.00 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.56 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc-rocwmma__hblt0__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        166.11 ± 0.32 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        169.35 ± 0.56 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         19.83 ± 0.00 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.65 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |         89.60 ± 0.20 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |         94.73 ± 0.22 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.22 ± 0.00 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.47 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |         64.66 ± 0.16 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |         93.27 ± 0.18 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.35 ± 0.00 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.67 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc__hblt0.log

@@ -2,5 +2,9 @@ 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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
-HW Exception by GPU node-1 (Agent handle: 0x1d380ea0) reason :GPU Hang
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
-✖ ! [rocm7_rc] GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002 __hblt0 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        159.89 ± 0.44 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         20.55 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__rocm7_rc__hblt0__fa1.log

@@ -2,5 +2,9 @@ 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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
-Memory access fault by GPU node-1 (Agent handle: 0x4a0fea0) on address 0x7f3bf796f000. Reason: Page not present or supervisor privilege.
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
-✖ ! [rocm7_rc] GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002 __hblt0__fa1 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        170.42 ± 0.33 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         20.66 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__vulkan_amdvlk.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (AMD open-source driver) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 32768 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |    0 |           pp512 |        197.95 ± 0.29 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |    0 |           pp512 |        217.22 ± 0.49 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |    0 |           tg128 |         23.24 ± 0.00 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |    0 |           tg128 |         24.18 ± 0.01 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__vulkan_amdvlk__fa1.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (AMD open-source driver) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 32768 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           pp512 |        199.40 ± 0.35 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           pp512 |        219.61 ± 0.55 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           tg128 |         23.26 ± 0.00 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           tg128 |         24.21 ± 0.01 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__vulkan_radv.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |    0 |           pp512 |        126.28 ± 0.17 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |    0 |           pp512 |        212.60 ± 0.74 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |    0 |           tg128 |         23.33 ± 0.01 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |    0 |           tg128 |         24.18 ± 0.03 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q4_K_XL-00001-of-00002__vulkan_radv__fa1.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           pp512 |        131.64 ± 0.32 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           pp512 |        224.85 ± 2.55 |
-| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           tg128 |         23.88 ± 0.01 |
+| glm4moe 106B.A12B Q4_K - Medium |  68.01 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           tg128 |         24.64 ± 0.01 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm6_4_3-rocwmma.log

@@ -9,7 +9,7 @@ rocBLAS warning: hipBlasLT failed, falling back to tensile.
 This message will be only be displayed once, unless the ROCBLAS_VERBOSE_TENSILE_ERROR environment variable is set.
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        121.82 ± 0.35 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        120.87 ± 0.23 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.59 ± 0.00 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.86 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm6_4_3-rocwmma__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        126.60 ± 0.30 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        128.65 ± 0.59 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.62 ± 0.04 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.96 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm6_4_3-rocwmma__hblt0.log

@@ -2,5 +2,5 @@ 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
-Memory access fault by GPU node-1 (Agent handle: 0x26e36340) on address 0x7fcef3635000. Reason: Page not present or supervisor privilege.
+HW Exception by GPU node-1 (Agent handle: 0xe6e7340) reason :GPU Hang
 ✖ ! [rocm6_4_3-rocwmma] GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003 __hblt0 failed (exit 134)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm6_4_3-rocwmma__hblt0__fa1.log

@@ -2,5 +2,5 @@ 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
-HW Exception by GPU node-1 (Agent handle: 0x35263340) reason :GPU Hang
+Memory access fault by GPU node-1 (Agent handle: 0x400a9340) on address 0x7ef17b435000. Reason: Page not present or supervisor privilege.
 ✖ ! [rocm6_4_3-rocwmma] GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003 __hblt0__fa1 failed (exit 134)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm6_4_3.log

@@ -9,7 +9,7 @@ rocBLAS warning: hipBlasLT failed, falling back to tensile.
 This message will be only be displayed once, unless the ROCBLAS_VERBOSE_TENSILE_ERROR environment variable is set.
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        117.95 ± 0.30 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        120.53 ± 0.28 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.65 ± 0.01 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.87 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm6_4_3__fa1.log

@@ -2,5 +2,9 @@ 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
-Memory access fault by GPU node-1 (Agent handle: 0x28aa3340) on address 0x7fb93761b000. Reason: Page not present or supervisor privilege.
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
-✖ ! [rocm6_4_3] GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003 __fa1 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        129.22 ± 0.41 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.95 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm6_4_3__hblt0.log

@@ -2,5 +2,5 @@ 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
-HW Exception by GPU node-1 (Agent handle: 0x14d05340) reason :GPU Hang
+Memory access fault by GPU node-1 (Agent handle: 0x22558310) on address 0x7f7830fad000. Reason: Page not present or supervisor privilege.
 ✖ ! [rocm6_4_3] GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003 __hblt0 failed (exit 134)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm6_4_3__hblt0__fa1.log

@@ -2,5 +2,9 @@ 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
-HW Exception by GPU node-1 (Agent handle: 0x265e8340) reason :GPU Hang
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
-✖ ! [rocm6_4_3] GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003 __hblt0__fa1 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        128.68 ± 0.22 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.96 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc-rocwmma-fa_all_quants.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |         91.95 ± 0.25 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.76 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc-rocwmma-fa_all_quants__fa1.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |         70.00 ± 0.17 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.98 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc-rocwmma-fa_all_quants__hblt0.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        134.22 ± 0.50 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.90 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc-rocwmma-fa_all_quants__hblt0__fa1.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        159.75 ± 0.33 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.99 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc-rocwmma.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |         69.19 ± 0.20 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |         92.18 ± 0.04 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.64 ± 0.00 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.92 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc-rocwmma__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        114.61 ± 0.20 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        121.75 ± 0.32 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.51 ± 0.00 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.97 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc-rocwmma__hblt0.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        120.88 ± 0.92 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        151.32 ± 0.45 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.61 ± 0.09 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.90 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc-rocwmma__hblt0__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        150.07 ± 0.56 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        161.10 ± 0.36 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.52 ± 0.00 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.99 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |         69.52 ± 0.17 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |         92.20 ± 0.11 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.63 ± 0.00 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.85 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |         74.02 ± 0.13 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |         71.02 ± 0.16 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.73 ± 0.00 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.96 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc__hblt0.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        142.67 ± 0.75 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           pp512 |        147.32 ± 0.43 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.68 ± 0.00 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |    0 |           tg128 |         15.91 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__rocm7_rc__hblt0__fa1.log

@@ -2,5 +2,9 @@ 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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
-Memory access fault by GPU node-1 (Agent handle: 0x1c536ea0) on address 0x7f623b57e000. Reason: Page not present or supervisor privilege.
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
-✖ ! [rocm7_rc] GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003 __hblt0__fa1 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           pp512 |        161.37 ± 0.36 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | ROCm       |  99 |  1 |    0 |           tg128 |         15.99 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__vulkan_amdvlk.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (AMD open-source driver) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 32768 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |    0 |           pp512 |        219.81 ± 0.70 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |    0 |           pp512 |        264.50 ± 0.99 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |    0 |           tg128 |         16.80 ± 0.00 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |    0 |           tg128 |         17.27 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__vulkan_amdvlk__fa1.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (AMD open-source driver) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 32768 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           pp512 |        222.20 ± 0.63 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           pp512 |        267.86 ± 1.22 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           tg128 |         16.82 ± 0.01 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           tg128 |         17.28 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__vulkan_radv.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |    0 |           pp512 |        126.55 ± 0.40 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |    0 |           pp512 |        208.01 ± 0.73 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |    0 |           tg128 |         17.07 ± 0.01 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |    0 |           tg128 |         17.49 ± 0.02 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/GLM-4.5-Air-UD-Q6_K_XL-00001-of-00003__vulkan_radv__fa1.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           pp512 |        131.25 ± 0.50 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           pp512 |        221.63 ± 1.26 |
-| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           tg128 |         17.31 ± 0.00 |
+| glm4moe 106B.A12B Q6_K         |  94.57 GiB |   110.47 B | Vulkan     |  99 |  1 |    0 |           tg128 |         17.71 ± 0.01 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm6_4_3-rocwmma.log

@@ -7,9 +7,5 @@ This message will be only be displayed once, unless the ROCBLAS_VERBOSE_HIPBLASL
 
 rocBLAS warning: hipBlasLT failed, falling back to tensile. 
 This message will be only be displayed once, unless the ROCBLAS_VERBOSE_TENSILE_ERROR environment variable is set.
-| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
+HW Exception by GPU node-1 (Agent handle: 0x284c3340) reason :GPU Hang
-| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+✖ ! [rocm6_4_3-rocwmma] Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002 failed (exit 134)
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         98.02 ± 0.18 |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           tg128 |          2.77 ± 0.00 |
-
-build: 1fe00296 (6182)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm6_4_3-rocwmma__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |        101.83 ± 0.11 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |        101.82 ± 0.06 |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           tg128 |          2.77 ± 0.00 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           tg128 |          2.78 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm6_4_3-rocwmma__hblt0.log

@@ -2,5 +2,5 @@ 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
-HW Exception by GPU node-1 (Agent handle: 0x21da1340) reason :GPU Hang
+HW Exception by GPU node-1 (Agent handle: 0x7166340) reason :GPU Hang
 ✖ ! [rocm6_4_3-rocwmma] Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002 __hblt0 failed (exit 134)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm6_4_3-rocwmma__hblt0__fa1.log

@@ -2,5 +2,5 @@ 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
-HW Exception by GPU node-1 (Agent handle: 0x15ac2340) reason :GPU Hang
+HW Exception by GPU node-1 (Agent handle: 0x37f0e340) reason :GPU Hang
 ✖ ! [rocm6_4_3-rocwmma] Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002 __hblt0__fa1 failed (exit 134)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm6_4_3.log

@@ -9,7 +9,7 @@ rocBLAS warning: hipBlasLT failed, falling back to tensile.
 This message will be only be displayed once, unless the ROCBLAS_VERBOSE_TENSILE_ERROR environment variable is set.
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         97.13 ± 0.17 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         94.79 ± 0.14 |
 | llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           tg128 |          2.78 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm6_4_3__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |         80.42 ± 0.08 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |        104.62 ± 0.08 |
 | llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           tg128 |          2.78 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm6_4_3__hblt0.log

@@ -2,5 +2,5 @@ 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
-HW Exception by GPU node-1 (Agent handle: 0x2c1e5340) reason :GPU Hang
+HW Exception by GPU node-1 (Agent handle: 0x12cee310) reason :GPU Hang
 ✖ ! [rocm6_4_3] Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002 __hblt0 failed (exit 134)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm6_4_3__hblt0__fa1.log

@@ -2,5 +2,5 @@ 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
-Memory access fault by GPU node-1 (Agent handle: 0x3e536340) on address 0x7f9182f6f000. Reason: Page not present or supervisor privilege.
+Memory access fault by GPU node-1 (Agent handle: 0x367c310) on address 0x7fc07ad93000. Reason: Page not present or supervisor privilege.
 ✖ ! [rocm6_4_3] Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002 __hblt0__fa1 failed (exit 134)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         98.15 ± 0.16 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           tg128 |          2.77 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants__fa1.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |        102.79 ± 0.14 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           tg128 |          2.78 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants__hblt0.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         93.89 ± 0.22 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           tg128 |          2.78 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants__hblt0__fa1.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |         97.53 ± 0.17 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           tg128 |          2.78 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc-rocwmma.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         97.31 ± 0.20 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         97.42 ± 0.12 |
 | llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           tg128 |          2.78 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc-rocwmma__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |        100.85 ± 0.13 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |        101.56 ± 0.04 |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           tg128 |          2.77 ± 0.00 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           tg128 |          2.78 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc-rocwmma__hblt0.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         93.00 ± 0.22 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         92.02 ± 0.17 |
 | llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           tg128 |          2.78 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc-rocwmma__hblt0__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |         97.88 ± 0.09 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |         97.10 ± 0.17 |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           tg128 |          2.77 ± 0.00 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           tg128 |          2.78 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         99.41 ± 0.36 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         95.12 ± 0.17 |
 | llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           tg128 |          2.77 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc__fa1.log

@@ -2,5 +2,9 @@ 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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
-Memory access fault by GPU node-1 (Agent handle: 0x1f66bec0) on address 0x7f3e84b6f000. Reason: Page not present or supervisor privilege.
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
-✖ ! [rocm7_rc] Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002 __fa1 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |        103.16 ± 0.07 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           tg128 |          2.78 ± 0.01 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc__hblt0.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         94.06 ± 0.09 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           pp512 |         93.86 ± 0.18 |
 | llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |    0 |           tg128 |          2.78 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__rocm7_rc__hblt0__fa1.log

@@ -2,5 +2,9 @@ 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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
-Memory access fault by GPU node-1 (Agent handle: 0xac09ec0) on address 0x7f283f56f000. Reason: Page not present or supervisor privilege.
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
-✖ ! [rocm7_rc] Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002 __hblt0__fa1 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           pp512 |         95.87 ± 0.08 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | ROCm       |  99 |  1 |    0 |           tg128 |          2.78 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__vulkan_amdvlk.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (AMD open-source driver) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 32768 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |    0 |           pp512 |         98.03 ± 0.24 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |    0 |           pp512 |         97.72 ± 0.36 |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |    0 |           tg128 |          2.78 ± 0.00 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |    0 |           tg128 |          2.81 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__vulkan_amdvlk__fa1.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (AMD open-source driver) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 32768 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |  1 |    0 |           pp512 |         99.12 ± 0.25 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |  1 |    0 |           pp512 |         99.04 ± 0.31 |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |  1 |    0 |           tg128 |          2.77 ± 0.00 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |  1 |    0 |           tg128 |          2.80 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__vulkan_radv.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |    0 |           pp512 |         75.59 ± 0.28 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |    0 |           pp512 |         78.94 ± 0.51 |
 | llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |    0 |           tg128 |          2.78 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-3.3-70B-Instruct-UD-Q8_K_XL-00001-of-00002__vulkan_radv__fa1.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |  1 |    0 |           pp512 |         80.09 ± 0.38 |
+| llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |  1 |    0 |           pp512 |         80.90 ± 0.77 |
 | llama 70B Q8_0                 |  75.65 GiB |    70.55 B | Vulkan     |  99 |  1 |    0 |           tg128 |          2.78 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm6_4_3-rocwmma.log

@@ -7,5 +7,9 @@ This message will be only be displayed once, unless the ROCBLAS_VERBOSE_HIPBLASL
 
 rocBLAS warning: hipBlasLT failed, falling back to tensile. 
 This message will be only be displayed once, unless the ROCBLAS_VERBOSE_TENSILE_ERROR environment variable is set.
-Memory access fault by GPU node-1 (Agent handle: 0x1a840340) on address 0x7f3babb56000. Reason: Page not present or supervisor privilege.
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
-✖ ! [rocm6_4_3-rocwmma] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        265.76 ± 0.95 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         14.69 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm6_4_3-rocwmma__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        291.08 ± 1.26 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        289.14 ± 1.57 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         14.53 ± 0.00 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         14.64 ± 0.15 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm6_4_3-rocwmma__hblt0.log

@@ -2,9 +2,5 @@ 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 | mmap |            test |                  t/s |
+HW Exception by GPU node-1 (Agent handle: 0x24187340) reason :GPU Hang
-| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+✖ ! [rocm6_4_3-rocwmma] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 __hblt0 failed (exit 134)
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        134.19 ± 1.49 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         14.56 ± 0.01 |
-
-build: 1fe00296 (6182)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm6_4_3-rocwmma__hblt0__fa1.log

@@ -2,5 +2,5 @@ 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
-HW Exception by GPU node-1 (Agent handle: 0x1de78340) reason :GPU Hang
+HW Exception by GPU node-1 (Agent handle: 0x3da9340) reason :GPU Hang
 ✖ ! [rocm6_4_3-rocwmma] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 __hblt0__fa1 failed (exit 134)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm6_4_3.log

@@ -7,9 +7,5 @@ This message will be only be displayed once, unless the ROCBLAS_VERBOSE_HIPBLASL
 
 rocBLAS warning: hipBlasLT failed, falling back to tensile. 
 This message will be only be displayed once, unless the ROCBLAS_VERBOSE_TENSILE_ERROR environment variable is set.
-| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
+HW Exception by GPU node-1 (Agent handle: 0x11bc3310) reason :GPU Hang
-| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+✖ ! [rocm6_4_3] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 failed (exit 134)
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        270.28 ± 1.29 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         14.58 ± 0.03 |
-
-build: 1fe00296 (6182)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm6_4_3__fa1.log

@@ -2,5 +2,9 @@ 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
-Memory access fault by GPU node-1 (Agent handle: 0x2162b340) on address 0x7f500556f000. Reason: Page not present or supervisor privilege.
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
-✖ ! [rocm6_4_3] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 __fa1 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        291.67 ± 0.91 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         14.71 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm6_4_3__hblt0.log

@@ -2,5 +2,5 @@ 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
-HW Exception by GPU node-1 (Agent handle: 0xdacf340) reason :GPU Hang
+HW Exception by GPU node-1 (Agent handle: 0x8a0a310) reason :GPU Hang
 ✖ ! [rocm6_4_3] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 __hblt0 failed (exit 134)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm6_4_3__hblt0__fa1.log

@@ -2,5 +2,5 @@ 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
-HW Exception by GPU node-1 (Agent handle: 0x3dc00340) reason :GPU Hang
+HW Exception by GPU node-1 (Agent handle: 0x1ada6310) reason :GPU Hang
 ✖ ! [rocm6_4_3] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 __hblt0__fa1 failed (exit 134)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        276.44 ± 1.46 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         14.55 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants__fa1.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        292.67 ± 1.04 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         14.71 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants__hblt0.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        273.88 ± 1.14 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         14.70 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc-rocwmma-fa_all_quants__hblt0__fa1.log

@@ -0,0 +1,10 @@
+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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        284.81 ± 1.55 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         14.72 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc-rocwmma.log

@@ -2,5 +2,9 @@ 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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
-HW Exception by GPU node-1 (Agent handle: 0x3882bf60) reason :GPU Hang
+| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
-✖ ! [rocm7_rc-rocwmma] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        274.13 ± 0.84 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         14.71 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc-rocwmma__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        285.84 ± 9.41 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        292.92 ± 2.63 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         14.37 ± 0.00 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         14.71 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc-rocwmma__hblt0.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        273.97 ± 1.67 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        273.23 ± 1.35 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         14.57 ± 0.05 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         14.70 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc-rocwmma__hblt0__fa1.log

@@ -2,9 +2,5 @@ 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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
-| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
+HW Exception by GPU node-1 (Agent handle: 0x13c5d180) reason :GPU Hang
-| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+✖ ! [rocm7_rc-rocwmma] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 __hblt0__fa1 failed (exit 134)
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        285.26 ± 1.79 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         14.33 ± 0.03 |
-
-build: de219279 (6181)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc.log

@@ -2,9 +2,5 @@ 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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
-| model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
+Memory access fault by GPU node-1 (Agent handle: 0x381db160) on address 0x7f72baf68000. Reason: Page not present or supervisor privilege.
-| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
+✖ ! [rocm7_rc] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 failed (exit 134)
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        276.37 ± 1.65 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         14.57 ± 0.04 |
-
-build: de219279 (6181)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc__fa1.log

@@ -2,5 +2,5 @@ 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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
-Memory access fault by GPU node-1 (Agent handle: 0xa893ec0) on address 0x7f070a3a9000. Reason: Page not present or supervisor privilege.
+HW Exception by GPU node-1 (Agent handle: 0x34902180) reason :GPU Hang
 ✖ ! [rocm7_rc] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 __fa1 failed (exit 134)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc__hblt0.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        269.17 ± 0.99 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        274.52 ± 1.78 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         14.63 ± 0.01 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         14.70 ± 0.00 |
 
-build: de219279 (6181)
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__rocm7_rc__hblt0__fa1.log

@@ -2,5 +2,9 @@ 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: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
-Memory access fault by GPU node-1 (Agent handle: 0x1db86ec0) on address 0x7f2273f6f000. Reason: Page not present or supervisor privilege.
+| model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
-✖ ! [rocm7_rc] Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002 __hblt0__fa1 failed (exit 134)
+| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        287.04 ± 1.92 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         14.71 ± 0.00 |
+
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__vulkan_amdvlk.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (AMD open-source driver) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 32768 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |    0 |           pp512 |        242.07 ± 1.05 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |    0 |           pp512 |        224.02 ± 2.86 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |    0 |           tg128 |         15.56 ± 0.01 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |    0 |           tg128 |         15.98 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__vulkan_amdvlk__fa1.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (AMD open-source driver) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 32768 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |  1 |    0 |           pp512 |        244.49 ± 1.13 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |  1 |    0 |           pp512 |        234.30 ± 1.10 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |  1 |    0 |           tg128 |         15.33 ± 0.00 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |  1 |    0 |           tg128 |         15.75 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__vulkan_radv.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |    0 |           pp512 |        147.08 ± 0.98 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |    0 |           pp512 |        201.49 ± 2.22 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |    0 |           tg128 |         15.50 ± 0.01 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |    0 |           tg128 |         15.77 ± 0.01 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q6_K-00001-of-00002__vulkan_radv__fa1.log

@@ -2,7 +2,7 @@ ggml_vulkan: Found 1 Vulkan devices:
 ggml_vulkan: 0 = Radeon 8060S Graphics (RADV GFX1151) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 1 | matrix cores: KHR_coopmat
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |  1 |    0 |           pp512 |        149.97 ± 1.10 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |  1 |    0 |           pp512 |        202.49 ± 5.98 |
-| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |  1 |    0 |           tg128 |         15.49 ± 0.00 |
+| llama4 17Bx16E (Scout) Q6_K    |  82.35 GiB |   107.77 B | Vulkan     |  99 |  1 |    0 |           tg128 |         15.74 ± 0.00 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q8_0-00001-of-00003__rocm6_4_3-rocwmma.log

@@ -9,7 +9,7 @@ rocBLAS warning: hipBlasLT failed, falling back to tensile.
 This message will be only be displayed once, unless the ROCBLAS_VERBOSE_TENSILE_ERROR environment variable is set.
 | model                          |       size |     params | backend    | ngl | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
-| llama4 17Bx16E (Scout) Q8_0    | 106.65 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |        270.35 ± 3.39 |
+| llama4 17Bx16E (Scout) Q8_0    | 106.65 GiB |   107.77 B | ROCm       |  99 |    0 |           pp512 |       264.44 ± 24.69 |
-| llama4 17Bx16E (Scout) Q8_0    | 106.65 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         11.78 ± 0.03 |
+| llama4 17Bx16E (Scout) Q8_0    | 106.65 GiB |   107.77 B | ROCm       |  99 |    0 |           tg128 |         11.88 ± 0.05 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q8_0-00001-of-00003__rocm6_4_3-rocwmma__fa1.log

@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
   Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
 | model                          |       size |     params | backend    | ngl | fa | mmap |            test |                  t/s |
 | ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
-| llama4 17Bx16E (Scout) Q8_0    | 106.65 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        292.23 ± 3.13 |
+| llama4 17Bx16E (Scout) Q8_0    | 106.65 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        298.83 ± 1.59 |
-| llama4 17Bx16E (Scout) Q8_0    | 106.65 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         11.73 ± 0.03 |
+| llama4 17Bx16E (Scout) Q8_0    | 106.65 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         11.89 ± 0.06 |
 
-build: 1fe00296 (6182)
+build: f1fbffb5 (6486)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q8_0-00001-of-00003__rocm6_4_3-rocwmma__hblt0.log

@@ -2,5 +2,5 @@ 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
-HW Exception by GPU node-1 (Agent handle: 0x5f69340) reason :GPU Hang
+HW Exception by GPU node-1 (Agent handle: 0x3265f340) reason :GPU Hang
 ✖ ! [rocm6_4_3-rocwmma] Llama-4-Scout-17B-16E-Instruct-Q8_0-00001-of-00003 __hblt0 failed (exit 134)

benchmark/results/Llama-4-Scout-17B-16E-Instruct-Q8_0-00001-of-00003__rocm6_4_3-rocwmma__hblt0__fa1.log

@@ -2,9 +2,5 @@ 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 | fa | mmap |            test |                  t/s |
+HW Exception by GPU node-1 (Agent handle: 0x33cad340) reason :GPU Hang
-| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
+✖ ! [rocm6_4_3-rocwmma] Llama-4-Scout-17B-16E-Instruct-Q8_0-00001-of-00003 __hblt0__fa1 failed (exit 134)
-| llama4 17Bx16E (Scout) Q8_0    | 106.65 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           pp512 |        140.27 ± 0.97 |
-| llama4 17Bx16E (Scout) Q8_0    | 106.65 GiB |   107.77 B | ROCm       |  99 |  1 |    0 |           tg128 |         11.74 ± 0.00 |
-
-build: 1fe00296 (6182)