Skip to content

llama-quantize: fix tensor-type logic #21482

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

Closed
theo77186 wants to merge 2 commits into from
Closed

llama-quantize: fix tensor-type logic #21482

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a9a1d27
llama-quantize: fix tensor-type logic
theo77186 Apr 5, 2026
3f53ec2
Remove stale manual variable assignment
theo77186 Apr 5, 2026
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
12 changes: 4 additions & 8 deletions src/llama-quant.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -673,8 +673,10 @@ static ggml_type llama_tensor_get_type(quantize_state_impl & qs, const llama_mod

// get more optimal quantization type based on the tensor shape, layer, etc.
if (!params->pure && ggml_is_quantized(default_type)) {
// if the user provided tensor types - use those
bool manual = false;
// use the standard logic for choosing the quantization type based on the selected mixture
new_type = llama_tensor_get_type_impl(qs, new_type, tensor, params->ftype, tm.category);

// if the user provided tensor types - use those instead
if (!qs.tensor_type_patterns.empty()) {
const std::string tensor_name(tensor->name);
for (const auto & [pattern, qtype] : qs.tensor_type_patterns) {
Expand All @@ -683,18 +685,12 @@ static ggml_type llama_tensor_get_type(quantize_state_impl & qs, const llama_mod
LLAMA_LOG_WARN("%s: %-36s - applying manual override: %s -> %s\n",
__func__, tensor_name.c_str(), ggml_type_name(new_type), ggml_type_name(qtype));
new_type = qtype;
manual = true;
break;
}
}
}
}

// if not manual - use the standard logic for choosing the quantization type based on the selected mixture
if (!manual) {
new_type = llama_tensor_get_type_impl(qs, new_type, tensor, params->ftype, tm.category);
}

// incompatible tensor shapes are handled here - fallback to a compatible type
new_type = tensor_type_fallback(qs, tensor, new_type);
}
Expand Down