From 61221221d77fa4a9823892eae2829b817f9e33f7 Mon Sep 17 00:00:00 2001 From: Herman Semenov Date: Mon, 26 Aug 2024 03:18:23 +0300 Subject: [PATCH] llama: changed default type IQ2_XS to IQ2_S for LLAMA_FTYPE_MOSTLY_IQ2_S --- src/llama.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/llama.cpp b/src/llama.cpp index fc8fb3e0d..3f173cfb5 100644 --- a/src/llama.cpp +++ b/src/llama.cpp @@ -16719,7 +16719,7 @@ static void llama_model_quantize_internal(const std::string & fname_inp, const s case LLAMA_FTYPE_MOSTLY_Q6_K: default_type = GGML_TYPE_Q6_K; break; case LLAMA_FTYPE_MOSTLY_IQ2_XXS: default_type = GGML_TYPE_IQ2_XXS; break; case LLAMA_FTYPE_MOSTLY_IQ2_XS: default_type = GGML_TYPE_IQ2_XS; break; - case LLAMA_FTYPE_MOSTLY_IQ2_S: default_type = GGML_TYPE_IQ2_XS; break; + case LLAMA_FTYPE_MOSTLY_IQ2_S: default_type = GGML_TYPE_IQ2_S; break; case LLAMA_FTYPE_MOSTLY_IQ2_M: default_type = GGML_TYPE_IQ2_S; break; case LLAMA_FTYPE_MOSTLY_IQ3_XXS: default_type = GGML_TYPE_IQ3_XXS; break; case LLAMA_FTYPE_MOSTLY_IQ1_S: default_type = GGML_TYPE_IQ1_S; break;