ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

mtmd : support InternVL 2.5 and 3 (#13422)

Browse source 
* convert : internvl support

* InternVL3-1B working

* fix regression

* rm mobilevlm from test

* fix conversion

* add test for internvl

* add to list of pre-quant

* restore boi/eoi check

* add clarify comment for norm eps
This commit is contained in:
Xuan-Son Nguyen 2025-05-10 16:26:42 +02:00 • committed by GitHub
parent d8919424f1
commit 053367d149
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
9 changed files with 243 additions and 25 deletions
Download patch file Download diff file Expand all files Collapse all files

12
gguf-py/gguf/tensor_mapping.py
View file

@ -905,6 +905,7 @@ class TensorNameMap:
MODEL_TENSOR.V_MMPROJ_MLP: (
"model.mm_projector.mlp.mlp.{bid}",
"mlp1.{bid}", # InternVL
),
MODEL_TENSOR.V_MMPROJ_PEG: (
@ -955,6 +956,7 @@ class TensorNameMap:
MODEL_TENSOR.V_ENC_INPUT_NORM: (
"vision_tower.vision_model.encoder.layers.{bid}.layer_norm1",
"vision_tower.vision_model.encoder.layers.{bid}.norm1", # InternVL
"vpm.encoder.layers.{bid}.layer_norm1",
"model.vision_model.encoder.layers.{bid}.layer_norm1", # SmolVLM
"vision_tower.transformer.layers.{bid}.attention_norm", # pixtral
@ -963,6 +965,7 @@ class TensorNameMap:
MODEL_TENSOR.V_ENC_OUTPUT: (
"vision_tower.vision_model.encoder.layers.{bid}.self_attn.out_proj",
"vision_tower.vision_model.encoder.layers.{bid}.attn.proj", # InternVL
"vpm.encoder.layers.{bid}.self_attn.out_proj",
"model.vision_model.encoder.layers.{bid}.self_attn.out_proj", # SmolVLM
"vision_tower.transformer.layers.{bid}.attention.o_proj", # pixtral
@ -971,6 +974,7 @@ class TensorNameMap:
MODEL_TENSOR.V_ENC_OUTPUT_NORM: (
"vision_tower.vision_model.encoder.layers.{bid}.layer_norm2",
"vision_tower.vision_model.encoder.layers.{bid}.norm2", # InternVL
"vpm.encoder.layers.{bid}.layer_norm2",
"model.vision_model.encoder.layers.{bid}.layer_norm2", # SmolVLM
"vision_tower.transformer.layers.{bid}.ffn_norm", # pixtral
@ -1000,6 +1004,14 @@ class TensorNameMap:
"visual.blocks.{bid}.mlp.down_proj", # qwen2.5vl
),
MODEL_TENSOR.V_LAYER_SCALE_1: (
"vision_tower.vision_model.encoder.layers.{bid}.ls1", # InternVL
),
MODEL_TENSOR.V_LAYER_SCALE_2: (
"vision_tower.vision_model.encoder.layers.{bid}.ls2", # InternVL
),
MODEL_TENSOR.V_PRE_NORM: (
"vision_tower.vision_model.pre_layrnorm",
"vision_tower.ln_pre", # pixtral