From f0b14e8c6957bee69445b59704ebdc95d22363ed Mon Sep 17 00:00:00 2001 From: qwopqwop200 Date: Thu, 13 Apr 2023 14:53:23 +0900 Subject: [PATCH] add Q4_2 --- llama.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/llama.h b/llama.h index 7a258a1e1..890b111ba 100644 --- a/llama.h +++ b/llama.h @@ -72,6 +72,8 @@ extern "C" { LLAMA_FTYPE_MOSTLY_Q4_0 = 2, // except 1d tensors LLAMA_FTYPE_MOSTLY_Q4_1 = 3, // except 1d tensors LLAMA_FTYPE_MOSTLY_Q4_1_SOME_F16 = 4, // tok_embeddings.weight and output.weight are F16 + LLAMA_FTYPE_MOSTLY_Q4_2 = 5, // except 1d tensors + LLAMA_FTYPE_MOSTLY_Q4_2_SOME_F16 = 6, // except 1d tensors }; LLAMA_API struct llama_context_params llama_context_default_params();