public_git/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/whisper.cpp.git synced 2026-10-07 15:01:16 +02:00
Code Issues Packages Projects Releases Wiki Activity

talk-llama : sync llama.cpp

Browse Source
This commit is contained in:
Georgi Gerganov
2026-02-08 09:29:10 +02:00
parent b0e81c1a2e
commit 4b23ff249e
26 changed files with 1661 additions and 179 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/talk-llama/llama-quant.cpp
+2 -2
View File
@@ -787,9 +787,9 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::
quantize &= name != LLM_TN(model.arch)(LLM_TENSOR_POS_EMBD, "weight");
quantize &= name != LLM_TN(model.arch)(LLM_TENSOR_TOKEN_TYPES, "weight");
// do not quantize Mamba's small yet 2D weights
// do not quantize Mamba /Kimi's small conv1d weights
// NOTE: can't use LLM_TN here because the layer number is not known
quantize &= name.find("ssm_conv1d.weight") == std::string::npos;
quantize &= name.find("ssm_conv1d") == std::string::npos;
quantize &= name.find("shortconv.conv.weight") == std::string::npos;
// do not quantize RWKV's small yet 2D weights