diff options
author | Johannes Gäßler <johannesg@5d6.de> | 2023-07-05 14:19:42 +0200 |
---|---|---|
committer | GitHub <noreply@github.com> | 2023-07-05 14:19:42 +0200 |
commit | 924dd22fd3ba93e097f8d19ba5cda919ca2fe2fb (patch) | |
tree | ca169c258f2d00f7e31c8b743a9f1206280b4d6b /README.md | |
parent | 051c70dcd55709c9cbbfa849af035951fe720433 (diff) |
Quantized dot products for CUDA mul mat vec (#2067)
Diffstat (limited to 'README.md')
-rw-r--r-- | README.md | 3 |
1 files changed, 2 insertions, 1 deletions
@@ -345,8 +345,9 @@ Building the program with BLAS support may lead to some performance improvements | Option | Legal values | Default | Description | |-------------------------|------------------------|---------|-------------| + | LLAMA_CUDA_FORCE_DMMV | Boolean | false | Force the use of dequantization + matrix vector multiplication kernels instead of using kernels that do matrix vector multiplication on quantized data. By default the decision is made based on compute capability (MMVQ for 7.0/Turing/RTX 2000 or higher). Does not affect k-quants. | | LLAMA_CUDA_DMMV_X | Positive integer >= 32 | 32 | Number of values in x direction processed by the CUDA dequantization + matrix vector multiplication kernel per iteration. Increasing this value can improve performance on fast GPUs. Power of 2 heavily recommended. Does not affect k-quants. | - | LLAMA_CUDA_DMMV_Y | Positive integer | 1 | Block size in y direction for the CUDA dequantization + mul mat vec kernels. Increasing this value can improve performance on fast GPUs. Power of 2 recommended. Does not affect k-quants. | + | LLAMA_CUDA_MMV_Y | Positive integer | 1 | Block size in y direction for the CUDA mul mat vec kernels. Increasing this value can improve performance on fast GPUs. Power of 2 recommended. Does not affect k-quants. | | LLAMA_CUDA_DMMV_F16 | Boolean | false | If enabled, use half-precision floating point arithmetic for the CUDA dequantization + mul mat vec kernels. Can improve performance on relatively recent GPUs. | | LLAMA_CUDA_KQUANTS_ITER | 1 or 2 | 2 | Number of values processed per iteration and per CUDA thread for Q2_K and Q6_K quantization formats. Setting this value to 1 can improve performance for slow GPUs. | |