tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-08-19 14:31:06 -04:00
Code Issues Packages Projects Releases Wiki Activity

llama : Add support for DeepSeek V3 (#11049)

Browse Source 
* convert : extend DEEPSEEK2 model architecture to support DeepseekV3ForCausalLM by adding EXPERT_WEIGHTS_NORM and EXPERT_GATING_FUNC model parameters and FFN_EXP_PROBS_B tensor type

* vocab : add DeepSeek V3 pre-tokenizer regexes

* unicode : handle ACCENT_MARK and SYMBOL categories in regex

* llama : add DeepSeek V3 chat template, handle new model parameters and tensor types

---------

Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
This commit is contained in:
fairydreaming
2025-01-04 21:06:11 +01:00
committed by GitHub
parent f922a9c542
commit 9394bbd484
16 changed files with 162 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
gguf-py/gguf/gguf_writer.py
View File

@@ -26,6 +26,7 @@ from .constants import (
RopeScalingType,
PoolingType,
TokenType,
ExpertGatingFuncType,
)
from .quants import quant_shape_from_byte_shape
@@ -715,6 +716,12 @@ class GGUFWriter:
def add_expert_weights_scale(self, value: float) -> None:
self.add_float32(Keys.LLM.EXPERT_WEIGHTS_SCALE.format(arch=self.arch), value)
def add_expert_weights_norm(self, value: bool) -> None:
self.add_bool(Keys.LLM.EXPERT_WEIGHTS_NORM.format(arch=self.arch), value)
def add_expert_gating_func(self, value: ExpertGatingFuncType) -> None:
self.add_uint32(Keys.LLM.EXPERT_GATING_FUNC.format(arch=self.arch), value.value)
def add_swin_norm(self, value: bool) -> None:
self.add_bool(Keys.LLM.SWIN_NORM.format(arch=self.arch), value)