From c85e139c68b0b0b15656cc0bc9618d632ed18822 Mon Sep 17 00:00:00 2001 From: Iwan Kawrakow Date: Sun, 28 Jul 2024 19:43:18 +0300 Subject: iq2_k: Basics Quantize/dequantize, CUDA deqantize, AVX512 iqk_mul_mat. --- ggml/include/ggml.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'ggml/include') diff --git a/ggml/include/ggml.h b/ggml/include/ggml.h index ff7f0064..2cb4af32 100644 --- a/ggml/include/ggml.h +++ b/ggml/include/ggml.h @@ -390,6 +390,7 @@ extern "C" { GGML_TYPE_IQ2_BN = 35, GGML_TYPE_Q8_K64 = 36, GGML_TYPE_IQ4_K = 37, + GGML_TYPE_IQ2_K = 38, GGML_TYPE_COUNT, }; @@ -437,6 +438,7 @@ extern "C" { GGML_FTYPE_MOSTLY_IQ1_BN = 28, // except 1d tensors GGML_FTYPE_MOSTLY_IQ2_BN = 29, // except 1d tensors GGML_FTYPE_MOSTLY_IQ4_K = 30, // except 1d tensors + GGML_FTYPE_MOSTLY_IQ2_K = 31, // except 1d tensors }; // available tensor operations: -- cgit v1.2.3