sleepy/llama.cpp
1
0
Fork 0
Code Issues 9 Pull Requests Actions 103 Packages Projects Releases Wiki Activity

model : Plamo3 support (#17304)

Browse Source 
* plamo3

* fix plamo3

* clean code

* clean up the code

* fix diff

* clean up the code

* clean up the code

* clean up the code

* clean up the code

* clean up the code

* clean up the code

* add chat_template if exist

* clean up the code

* fix cpu-backend

* chore: whitespace trim fix + typo fix

* Fix: address review feedback

* restore `FREQ_BASE_SWA` constant

* Fix: address review feedback2

* Fix:typecheck

* Fix: address review feedback3

* final cleanup

---------

Co-authored-by: mmngays <146910567+mmngays@users.noreply.github.com>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
momonga
2025-12-29 01:28:31 +09:00
committed by GitHub
parent 07a0c4ba92
commit 9c675c7140
9 changed files with 367 additions and 81 deletions
Download Patch File Download Diff File Expand all files Collapse all files
gguf-py/gguf/constants.py
+17
View File
@@ -377,6 +377,7 @@ class MODEL_ARCH(IntEnum):
PHIMOE = auto()
PLAMO = auto()
PLAMO2 = auto()
PLAMO3 = auto()
CODESHELL = auto()
ORION = auto()
INTERNLM2 = auto()
@@ -773,6 +774,7 @@ MODEL_ARCH_NAMES: dict[MODEL_ARCH, str] = {
MODEL_ARCH.PHIMOE: "phimoe",
MODEL_ARCH.PLAMO: "plamo",
MODEL_ARCH.PLAMO2: "plamo2",
MODEL_ARCH.PLAMO3: "plamo3",
MODEL_ARCH.CODESHELL: "codeshell",
MODEL_ARCH.ORION: "orion",
MODEL_ARCH.INTERNLM2: "internlm2",
@@ -1763,6 +1765,21 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.SSM_B_NORM,
MODEL_TENSOR.SSM_C_NORM,
],
MODEL_ARCH.PLAMO3: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.OUTPUT_NORM,
MODEL_TENSOR.OUTPUT,
MODEL_TENSOR.ATTN_NORM,
MODEL_TENSOR.ATTN_QKV,
MODEL_TENSOR.ATTN_Q_NORM,
MODEL_TENSOR.ATTN_K_NORM,
MODEL_TENSOR.ATTN_OUT,
MODEL_TENSOR.ATTN_POST_NORM,
MODEL_TENSOR.FFN_NORM,
MODEL_TENSOR.FFN_DOWN,
MODEL_TENSOR.FFN_UP,
MODEL_TENSOR.FFN_POST_NORM,
],
MODEL_ARCH.GPT2: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.POS_EMBD,
gguf-py/gguf/tensor_mapping.py
+2
View File
@@ -595,6 +595,7 @@ class TensorNameMap:
"encoder.layer.{bid}.attention.self.layer_norm_q", # jina-bert-v2
"transformer.layers.{bid}.attn.q_norm", # openelm
"model.layers.layers.{bid}.mixer.q", # plamo2
"model.layers.layers.{bid}.mixer.q_norm", # plamo3
"layers.{bid}.self_attn.q_norm", # qwen3-embedding
"model.layers.{bid}.attention.query_layernorm", # apertus
),
@@ -610,6 +611,7 @@ class TensorNameMap:
"encoder.layer.{bid}.attention.self.layer_norm_k", # jina-bert-v2
"transformer.layers.{bid}.attn.k_norm", # openelm
"model.layers.layers.{bid}.mixer.k", # plamo2
"model.layers.layers.{bid}.mixer.k_norm", # plamo3
"layers.{bid}.self_attn.k_norm", # qwen3-embedding
"model.layers.{bid}.attention.key_layernorm", # apertus
),