]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
ggml-cpu: add 128-bit RVV implementation for Quantization Vector Dot (llama/20633)
authorrehan-10xengineer <redacted>
Thu, 16 Apr 2026 08:15:15 +0000 (13:15 +0500)
committerGeorgi Gerganov <redacted>
Thu, 30 Apr 2026 08:29:11 +0000 (11:29 +0300)
commit94d6d0b743206b10a3074ea805c385b18fcd1498
tree35b94b0119151907014cf1cdceb81ca523cb8a7f
parent07c181b57f6d59027ea6fe3931967993e4f870a6
ggml-cpu: add 128-bit RVV implementation for Quantization Vector Dot (llama/20633)

* ggml-cpu: add 128-bit impls for i-quants, ternary quants

* ggml-cpu: add 128-bit impls for iq2_xs, iq3_s, iq3_xxs, tq2_0

Co-authored-by: Rehan Qasim <redacted>
* ggml-cpu: refactor; add rvv checks

---------

Co-authored-by: taimur-10x <redacted>
Co-authored-by: Rehan Qasim <redacted>
ggml/src/ggml-cpu/arch/riscv/quants.c