]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
opencl: add MoE support for q4_k, q5_k, q6_k on Adreno (llama/23303)
authorshaofeiqi <redacted>
Tue, 19 May 2026 21:29:00 +0000 (14:29 -0700)
committerGeorgi Gerganov <redacted>
Mon, 25 May 2026 09:26:07 +0000 (12:26 +0300)
commit37f17208c25aae5c93177599c05de6efc3b57446
treee878b9c8a2c66dcd8dc709c385f8e6455f6bec51
parentaca63e76386b239cbd1692629ddece4e7e26f03b
opencl: add MoE support for q4_k, q5_k, q6_k on Adreno (llama/23303)

* opencl: add q4_k moe support

* opencl: add q5_k moe support

* opencl: add q6_k moe support

* opencl: adjust format

---------

Co-authored-by: Li He <redacted>
ggml/src/ggml-opencl/CMakeLists.txt
ggml/src/ggml-opencl/ggml-opencl.cpp
ggml/src/ggml-opencl/kernels/cvt.cl
ggml/src/ggml-opencl/kernels/gemm_moe_q4_k_f32_ns.cl [new file with mode: 0644]
ggml/src/ggml-opencl/kernels/gemm_moe_q5_k_f32_ns.cl [new file with mode: 0644]
ggml/src/ggml-opencl/kernels/gemm_moe_q6_k_f32_ns.cl [new file with mode: 0644]
ggml/src/ggml-opencl/kernels/gemv_moe_q4_k_f32_ns.cl [new file with mode: 0644]
ggml/src/ggml-opencl/kernels/gemv_moe_q5_k_f32_ns.cl [new file with mode: 0644]
ggml/src/ggml-opencl/kernels/gemv_moe_q6_k_f32_ns.cl [new file with mode: 0644]