Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Static quant fallback improvements #4

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jun 7, 2025
Merged

Static quant fallback improvements #4

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 22 additions & 2 deletions src/llama-quant.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -846,12 +846,32 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::
// since many people will miss the error and not realize that most of the model is being quantized without an imatrix
// tok_embd should be ignored in this case, since it always causes this warning
if (name != tn(LLM_TENSOR_TOKEN_EMBD, "weight")) {
throw std::runtime_error(format("imatrix size %d is different from tensor size %d for %s",
int(it->second.size()), int(tensor->ne[0]*tensor->ne[2]), tensor->name));
LLAMA_LOG_INFO("imatrix size %d is different from tensor size %d for %s", int(it->second.size()), int(tensor->ne[0]*tensor->ne[2]), tensor->name);
}
}
}
}
if ((new_type == GGML_TYPE_IQ2_XXS ||
new_type == GGML_TYPE_IQ1_S ||
(new_type == GGML_TYPE_IQ1_M && strcmp(tensor->name, "token_embd.weight") && strcmp(tensor->name, "output.weight"))) && !imatrix) {
LLAMA_LOG_INFO("\n\n============================================================\n");
LLAMA_LOG_INFO("Missing importance matrix for tensor %s in a very low-bit quantization\n", tensor->name);
LLAMA_LOG_INFO("The result will be garbage, so using GGML_TYPE_Q2_K\n");
LLAMA_LOG_INFO("============================================================\n\n");
new_type = GGML_TYPE_Q2_K;
//throw std::runtime_error(format("Missing importance matrix for tensor %s in a very low-bit quantization", tensor->name));
}

if ((new_type == GGML_TYPE_IQ2_XS ||
new_type == GGML_TYPE_IQ2_S ||
(new_type == GGML_TYPE_Q2_K && params->ftype == LLAMA_FTYPE_MOSTLY_Q2_K_S && strcmp(tensor->name, "token_embd.weight") != 0)) && !imatrix) {
LLAMA_LOG_INFO("\n\n============================================================\n");
LLAMA_LOG_INFO("Missing importance matrix for tensor %s in a very low-bit quantization\n", tensor->name);
LLAMA_LOG_INFO("The result will be garbage, so using GGML_TYPE_Q3_K\n");
LLAMA_LOG_INFO("============================================================\n\n");
new_type = GGML_TYPE_Q3_K;
}

if ((new_type == GGML_TYPE_IQ2_XXS ||
new_type == GGML_TYPE_IQ2_XS ||
new_type == GGML_TYPE_IQ2_S ||
Expand Down
Loading