Strata

0 Fork 0

mirror of https://github.com/Niko1221/Strata.git synced 2026-11-10 11:51:31 -10:00

Files

T

efeisot 68c573cee8 AMD: ship the gfx1200 hipBLASLt table, with the with/without measurements

The table #256 left out (489623c): tools/hip/gfx1200-hipblaslt-100202.txt, 67
rows calibrated on the RX 9060 XT with tools/hip/tune_hipblaslt (system ROCm
7.2, hipBLASLt 1.2.2). Same binary, prompt speed with the table vs plain
hipBLAS, two runs each, Coder IQ1_M greedy MTP, STRATA_PREFILL_MMQ=1:

| prompt | table | plain hipBLAS | gain |
| 2,374  | 597-599 | 396-398 | 1.50x |
| 65,045 | 754-757 | 380     | 1.99x |
| 130,091| 719     | 363-370 | 1.98x |

docs/AMD_HIP.md: the community-validated section carries the numbers; ctest
37/37 (without ple_parity and platform_memory_test).

Continues #256.

🤖 Generated with Codebuff
Co-Authored-By: Codebuff <[email protected]>

2026-10-02 22:03:36 +02:00

bench_prefill.py

AMD HIP backend for gfx1100 (PR #121 , squashed onto its 0.1.21 base)

2026-09-29 14:33:45 +02:00

build_windows.bat

tools/hip: build_windows.bat and package_windows.py - the ready-made Windows HIP engine zip

2026-10-01 20:33:37 +02:00

gfx1100-hipblaslt-100100.txt

AMD HIP backend for gfx1100 (PR #121 , squashed onto its 0.1.21 base)

2026-09-29 14:33:45 +02:00

gfx1100-hipblaslt-100200.txt

AMD: hipBLASLt table for 100200 (the pinned ROCm), used by setup's configs (+42-59% prompt speed on the 7900 XTX); server: a config's env reaches the engine