Skip to content
KernelIndex
Search⌘K

submission 13410

KernelBot user 334 · python · License unknown

Use it

Vendorable · source mirrored · license unknownView source →

No package. Vendor the mirrored source: 1 lines, June 9 Researcher Reciprocity License v1.0.

template-hip.py
curl "https://kernelindex.com/api/v1/implementations/kernelbot-amd-fp8-mm-13410?include=source"
interfacepython
Compatibility
measured onAMD Instinct MI300X
declared hardwareAMD Instinct MI300X
architecturesgfx942
dtypesfp32, fp8_e4m3fnuz

Benchmark evidence

18 measurements across 1 GPU, fastest first.

Operation / workload
Hardware
Latency
Rank
Observed
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 256]
AMD Instinct MI300X
597.3µs
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 512]
AMD Instinct MI300X
666.3µs
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 7168] · n=512 · seed=6563
AMD Instinct MI300X
1.27ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 1536]
AMD Instinct MI300X
1.37ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 7168] · n=576 · seed=12346
AMD Instinct MI300X
1.40ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 256]
AMD Instinct MI300X
3.16ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 512]
AMD Instinct MI300X
3.64ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 7168] · n=1536 · seed=8135
AMD Instinct MI300X
3.64ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 2048]
AMD Instinct MI300X
4.30ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 2304]
AMD Instinct MI300X
4.90ms
#98 of 98
2025-04-30
Show all 18 measurements ›
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 7168] · n=512 · seed=12341
AMD Instinct MI300X
8.06ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 7168] · n=576 · seed=9863
AMD Instinct MI300X
8.54ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 1536]
AMD Instinct MI300X
9.39ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 7168] · n=4608 · seed=7531
AMD Instinct MI300X
10.2ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 7168] · n=1536 · seed=6543
AMD Instinct MI300X
24.8ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 2048]
AMD Instinct MI300X
29.2ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 2304]
AMD Instinct MI300X
32.9ms
#98 of 98
2025-04-30
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 7168] · n=4608 · seed=65436
AMD Instinct MI300X
74.7ms
#98 of 98
2025-04-30

Reported · How evidence levels are derived →

Source and license

sourceavailable
revision digestsha256:85fcba379c35d6a6625fa6371d3fe6cdbd296f7140df3db6e177bdd57a37f189
license declaredunknown
license concludedunknown
imported2026-08-15

Kernel source

template-hip.py1 lines
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

Source code from GPU Mode and the KernelBot dataset · June 9 Researcher Reciprocity License v1.0

Best evidence level for this revision: reported

JSON