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

mtmd : support SmolVLM (version 1 and 2) (#13050)

Browse source 
* mtmd : support SmolVLM (version 1 and 2)

* correct chat template

* fix n_patches

* scale_factor is an int

* add more models to test
This commit is contained in:
Xuan-Son Nguyen 2025-04-22 16:24:54 +02:00 • committed by GitHub
parent ab47dec3d3
commit dc39a5e7a8
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
10 changed files with 279 additions and 65 deletions
Download patch file Download diff file Expand all files Collapse all files

4
gguf-py/gguf/tensor_mapping.py
View file

@ -961,13 +961,13 @@ class TensorNameMap:
MODEL_TENSOR.V_ENC_FFN_UP: (
"vision_tower.vision_model.encoder.layers.{bid}.mlp.fc1",
"vpm.encoder.layers.{bid}.mlp.fc1",
"model.vision_model.encoder.layers.{bid}.mlp.fc1", # SmolVLM
"model.vision_model.encoder.layers.{bid}.mlp.fc2", # SmolVLM, gemma3 (note: name is swapped)
),
MODEL_TENSOR.V_ENC_FFN_DOWN: (
"vision_tower.vision_model.encoder.layers.{bid}.mlp.fc2",
"vpm.encoder.layers.{bid}.mlp.fc2",
"model.vision_model.encoder.layers.{bid}.mlp.fc2", # SmolVLM
"model.vision_model.encoder.layers.{bid}.mlp.fc1", # SmolVLM, gemma3 (note: name is swapped)
),
MODEL_TENSOR.V_PRE_NORM: (