Skip to content
KernelIndex
Search⌘K

submission 3803

KernelBot user 160 · 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.

submission.py
curl "https://kernelindex.com/api/v1/implementations/kernelbot-amd-fp8-mm-3803?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, 512]
AMD Instinct MI300X
205.7µs
#92 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 256]
AMD Instinct MI300X
215.6µs
#92 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 1536]
AMD Instinct MI300X
291.7µs
#92 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 7168] · n=512 · seed=6563
AMD Instinct MI300X
354.6µs
#87 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 7168] · n=576 · seed=12346
AMD Instinct MI300X
389.0µs
#90 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 256]
AMD Instinct MI300X
518.4µs
#92 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 512]
AMD Instinct MI300X
525.5µs
#93 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 2048]
AMD Instinct MI300X
661.8µs
#93 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 2304]
AMD Instinct MI300X
723.2µs
#93 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 7168] · n=1536 · seed=8135
AMD Instinct MI300X
746.3µs
#94 of 98
2025-04-16
Show all 18 measurements ›
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 1536]
AMD Instinct MI300X
1.01ms
#93 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 7168] · n=512 · seed=12341
AMD Instinct MI300X
1.35ms
#93 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 7168] · n=576 · seed=9863
AMD Instinct MI300X
1.50ms
#93 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [1024, 7168] · n=4608 · seed=7531
AMD Instinct MI300X
1.54ms
#93 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 7168] · n=1536 · seed=6543
AMD Instinct MI300X
2.55ms
#91 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 2048]
AMD Instinct MI300X
2.63ms
#88 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 2304]
AMD Instinct MI300X
3.01ms
#88 of 98
2025-04-16
AMD FP8 blockwise GEMMfp32/fp8_e4m3fnuz · [6144, 7168] · n=4608 · seed=65436
AMD Instinct MI300X
6.11ms
#89 of 98
2025-04-16

Reported · How evidence levels are derived →

Source and license

sourceavailable
revision digestsha256:4562c95646ea66788fac848af0344362c62f29381d88bc160f47fabad1344c5e
license declaredunknown
license concludedunknown
imported2026-08-15

Kernel source

submission.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