CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-17 11:37:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

model : add AfmoeForCausalLM support (#16477)

Browse Source 
* Add AFMOE model support

* Update to vocab

* Add model sizing

* Undo Rope change for ARCEE model

* Address review comments

* Update modeling code is_sliding -> use_rope, replace hard-coded logic

* Fix AFMOE tokenizer

* Update convert_hf_to_gguf.py

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* Update convert_hf_to_gguf.py

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* Update AFMoE tokenizer class identification to be more unique

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
Bartowski
2025-11-14 07:54:10 -05:00
committed by GitHub
parent 6cd0cf72ce
commit e1fcf8b09b
14 changed files with 541 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

31
gguf-py/gguf/constants.py
View File

@@ -409,6 +409,7 @@ class MODEL_ARCH(IntEnum):
BAILINGMOE2 = auto()
DOTS1 = auto()
ARCEE = auto()
AFMOE = auto()
ERNIE4_5 = auto()
ERNIE4_5_MOE = auto()
HUNYUAN_MOE = auto()
@@ -464,6 +465,7 @@ class MODEL_TENSOR(IntEnum):
ATTN_POST_NORM = auto()
ATTN_ROT_EMBD = auto()
ATTN_SINKS = auto()
ATTN_GATE = auto()
FFN_GATE_INP = auto()
FFN_GATE_INP_SHEXP = auto()
FFN_NORM = auto()
@@ -776,6 +778,7 @@ MODEL_ARCH_NAMES: dict[MODEL_ARCH, str] = {
MODEL_ARCH.BAILINGMOE2: "bailingmoe2",
MODEL_ARCH.DOTS1: "dots1",
MODEL_ARCH.ARCEE: "arcee",
MODEL_ARCH.AFMOE: "afmoe",
MODEL_ARCH.ERNIE4_5: "ernie4_5",
MODEL_ARCH.ERNIE4_5_MOE: "ernie4_5-moe",
MODEL_ARCH.FALCON_H1: "falcon-h1",
@@ -828,6 +831,7 @@ TENSOR_NAMES: dict[MODEL_TENSOR, str] = {
MODEL_TENSOR.ATTN_OUT: "blk.{bid}.attn_output",
MODEL_TENSOR.ATTN_ROT_EMBD: "blk.{bid}.attn_rot_embd",
MODEL_TENSOR.ATTN_SINKS: "blk.{bid}.attn_sinks",
MODEL_TENSOR.ATTN_GATE: "blk.{bid}.attn_gate",
MODEL_TENSOR.ATTN_Q_NORM: "blk.{bid}.attn_q_norm",
MODEL_TENSOR.ATTN_K_NORM: "blk.{bid}.attn_k_norm",
MODEL_TENSOR.ATTN_OUT_NORM: "blk.{bid}.attn_output_norm",
@@ -2693,6 +2697,33 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.FFN_DOWN,
MODEL_TENSOR.FFN_UP,
],
MODEL_ARCH.AFMOE: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.OUTPUT_NORM,
MODEL_TENSOR.OUTPUT,
MODEL_TENSOR.ATTN_NORM,
MODEL_TENSOR.ATTN_POST_NORM,
MODEL_TENSOR.ATTN_Q,
MODEL_TENSOR.ATTN_K,
MODEL_TENSOR.ATTN_V,
MODEL_TENSOR.ATTN_OUT,
MODEL_TENSOR.ATTN_Q_NORM,
MODEL_TENSOR.ATTN_K_NORM,
MODEL_TENSOR.ATTN_GATE,
MODEL_TENSOR.FFN_GATE,
MODEL_TENSOR.FFN_DOWN,
MODEL_TENSOR.FFN_UP,
MODEL_TENSOR.FFN_GATE_INP,
MODEL_TENSOR.FFN_GATE_EXP,
MODEL_TENSOR.FFN_DOWN_EXP,
MODEL_TENSOR.FFN_UP_EXP,
MODEL_TENSOR.FFN_GATE_SHEXP,
MODEL_TENSOR.FFN_UP_SHEXP,
MODEL_TENSOR.FFN_DOWN_SHEXP,
MODEL_TENSOR.FFN_PRE_NORM,
MODEL_TENSOR.FFN_POST_NORM,
MODEL_TENSOR.FFN_EXP_PROBS_B,
],
MODEL_ARCH.ERNIE4_5: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.OUTPUT_NORM,