baalajimaestro/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Update llama.cpp

Browse source
This commit is contained in:
Andrei Betlen 2024-01-14 08:17:22 -05:00
parent 7c898d5684
commit 359ae73643
2 changed files with 5 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
llama_cpp/llama_cpp.py
View file

@ -526,6 +526,7 @@ It might not exist for progress report where '.' is output repeatedly."""
# bool quantize_output_tensor; // quantize output.weight # bool quantize_output_tensor; // quantize output.weight
# bool only_copy; // only copy tensors - ftype, allow_requantize and quantize_output_tensor are ignored # bool only_copy; // only copy tensors - ftype, allow_requantize and quantize_output_tensor are ignored
# bool pure; // disable k-quant mixtures and quantize all tensors to the same type # bool pure; // disable k-quant mixtures and quantize all tensors to the same type
# void * imatrix; // pointer to importance matrix data
# } llama_model_quantize_params; # } llama_model_quantize_params;
class llama_model_quantize_params(Structure): class llama_model_quantize_params(Structure):
"""Parameters for llama_model_quantize """Parameters for llama_model_quantize
@ -537,6 +538,7 @@ class llama_model_quantize_params(Structure):
quantize_output_tensor (bool): quantize output.weight quantize_output_tensor (bool): quantize output.weight
only_copy (bool): only copy tensors - ftype, allow_requantize and quantize_output_tensor are ignored only_copy (bool): only copy tensors - ftype, allow_requantize and quantize_output_tensor are ignored
pure (bool): disable k-quant mixtures and quantize all tensors to the same type pure (bool): disable k-quant mixtures and quantize all tensors to the same type
imatrix (ctypes.c_void_p): pointer to importance matrix data
""" """
_fields_ = [ _fields_ = [
@ -545,6 +547,8 @@ class llama_model_quantize_params(Structure):
("allow_requantize", c_bool), ("allow_requantize", c_bool),
("quantize_output_tensor", c_bool), ("quantize_output_tensor", c_bool),
("only_copy", c_bool), ("only_copy", c_bool),
("pure", c_bool),
("imatrix", c_void_p),
] ]

2
vendor/llama.cpp vendored

@ -1 +1 @@
Subproject commit 76484fbfd355df388f71d6edaa98e1692a74de7e Subproject commit bb0c1392479398f9aba86d9ec98db0b95ede6e6d