thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

model : wire up Nemotron-H tensors for NVFP4 support (#20561)

Browse source 
* wire up Nemotron-H tensors for NVFP4 support

* add ssm tensors

* alignment
This commit is contained in:
Sigbjørn Skjæret 2026-03-16 09:19:16 +01:00 • committed by GitHub
parent 079e5a45f0
commit de8f01c2d7
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 18 additions and 11 deletions
Download patch file Download diff file Expand all files Collapse all files

3
src/llama-model.cpp
View file

@ -7501,6 +7501,9 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
}
// recurrent / linear-attention weight scales (per-tensor, shape {1})
if (!layer.ssm_in_s && layer.ssm_in) {
layer.ssm_in_s = create_tensor(tn(LLM_TENSOR_SSM_IN, "scale", i), {1}, TENSOR_NOT_REQUIRED);
}
if (!layer.ssm_out_s && layer.ssm_out) {
layer.ssm_out_s = create_tensor(tn(LLM_TENSOR_SSM_OUT, "scale", i), {1}, TENSOR_NOT_REQUIRED);
}