]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
Add `--no-op-offload` to improve `-ot` pp perf in MoE models like llama4 400B (llama...
authorDavid Huang <redacted>
Sun, 11 May 2025 12:18:39 +0000 (20:18 +0800)
committerGeorgi Gerganov <redacted>
Tue, 13 May 2025 10:02:19 +0000 (13:02 +0300)
commit6c46cbe30ef5fd644771570293d8190cb32b7348
tree90a0a2ab25975dbe239f95bc378210a796f95786
parent38648430fce1422694f2f349a5fe60d5969d6f49
Add `--no-op-offload` to improve `-ot` pp perf in MoE models like llama4 400B (llama/13386)
include/ggml-backend.h
src/ggml-backend.cpp
tests/test-opt.cpp