CYBER WATCH / EQUINOXE INTELLIGENCE
← All articles
b12205: webgpu – add COUNT_EQUAL support (#31578)
llama.cpp Releases · 2026-11-08 16:01:08
The commit adds COUNT_EQUAL support to the webgpu backend, optimizes small input sets by handling them in a single dispatch, simplifies tests by reusing existing ones, moves the COUNT_EQUAL pipeline cache declaration, and updates ggml‑webgpu.cpp. The changes are co‑authored by Masashi Yoshimura.Read original source ↗