xrip e14d7024f7 sm75: record tests and memory results

2026-09-21 20:56:54 +03:00

..

profile-shared-chat.sh

sm75: record tests and memory results

2026-09-21 20:56:54 +03:00

run-shared-chat-kernels.py

cuda: add opt-in SM75 SwiGLU input fusion

2026-09-21 20:56:20 +03:00

run-shared-chat-model.py

sm75: record tests and memory results

2026-09-21 20:56:54 +03:00

run-shared-chat-mtp.py

sm75: record tests and memory results

2026-09-21 20:56:54 +03:00

run-shared-chat-trace.py

sm75: record tests and memory results

2026-09-21 20:56:54 +03:00

shared-chat-mtp-request.json

sm75: record tests and memory results

2026-09-21 20:56:54 +03:00

SHARED-CHAT-PERFORMANCE.md

sm75: record tests and memory results

2026-09-21 20:56:54 +03:00

SHARED-CHAT-TESTS.md

sm75: record tests and memory results

2026-09-21 20:56:54 +03:00

test-gdn-transaction.cpp

llama: add opt-in SM75 recurrent state replay

2026-09-21 20:56:53 +03:00

test-mtp-state.cpp

llama: add opt-in SM75 recurrent state replay

2026-09-21 20:56:53 +03:00

test-shared-compute.cpp

llama: share SM75 MTP compute buffers on request

2026-09-21 20:56:54 +03:00