diff options
| author | Iwan Kawrakow <iwan.kawrakow@gmail.com> | 2024-06-18 13:56:16 +0300 |
|---|---|---|
| committer | Iwan Kawrakow <iwan.kawrakow@gmail.com> | 2024-06-22 12:02:52 +0300 |
| commit | 01ea9a862d4afb73f936de8f4ef46401ce11b596 (patch) | |
| tree | 892dd475ddf47c9ba401859b285b4b517d36135e /ggml-cuda.cu | |
| parent | 2998ca9b14d9b2d4b184cf6d923cea8b07a6320a (diff) | |
Bitnet(2.25 bpw): CUDA
We get PP-512 = 9600 t/s, TG-128 = 234 t/s
(but we need to use 8 CPU threads, else results are lower,
so clearly there is something being computed on the CPU).
PP-512 is very close to PP-512(fp16) = 9800 t/s
Diffstat (limited to 'ggml-cuda.cu')
| -rw-r--r-- | ggml-cuda.cu | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/ggml-cuda.cu b/ggml-cuda.cu index a3874e20..d51663c8 100644 --- a/ggml-cuda.cu +++ b/ggml-cuda.cu @@ -2757,7 +2757,7 @@ GGML_CALL static bool ggml_backend_cuda_supports_op(ggml_backend_t backend, cons if (a_type == GGML_TYPE_IQ2_XXS || a_type == GGML_TYPE_IQ2_XS || a_type == GGML_TYPE_IQ3_XXS || a_type == GGML_TYPE_IQ1_S || a_type == GGML_TYPE_IQ4_NL || a_type == GGML_TYPE_IQ3_S || a_type == GGML_TYPE_IQ1_M || a_type == GGML_TYPE_IQ2_S || a_type == GGML_TYPE_IQ4_XS || - a_type == GGML_TYPE_IQ1_BN) { + a_type == GGML_TYPE_IQ1_BN || a_type == GGML_TYPE_IQ2_BN) { if (b->ne[1] == 1 && ggml_nrows(b) > 1) { return false; } |