llama.cpp
30322097 - llama : deny -sm tensor for arches with moe amax rescaling (#31081)

Commit
26 days ago
llama : deny -sm tensor for arches with moe amax rescaling (#31081) The amax POOL_1D reduction spans the whole matmul row, so a device cannot compute it from an axis-0 split. The Meta split-state case reported MIRRORED, which gives each device a partial max of its own slice. Drop that case and deny -sm tensor for the arches which build the amax graph, see [TAG_MOE_AMAX_SCALE]. Assisted-by: dsh:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)
Author
Parents
Loading