Visual Question Answering
Transformers
Safetensors
English
Chinese
qwen2
text-generation
multimodal
text-generation-inference
Instructions to use BAAI/Aquila-VL-2B-llava-qwen with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use BAAI/Aquila-VL-2B-llava-qwen with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("visual-question-answering", model="BAAI/Aquila-VL-2B-llava-qwen")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("BAAI/Aquila-VL-2B-llava-qwen") model = AutoModelForCausalLM.from_pretrained("BAAI/Aquila-VL-2B-llava-qwen", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update config.json
Browse files- config.json +1 -2
config.json
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "/share/project/gushuhao/1.research/LLaVA-NeXT-main/llavaonevision-qwen1.5b-mlp2x_gelu-qwen2.5-stage1.5b-1012-32n40g-node8",
|
| 3 |
"architectures": [
|
| 4 |
"LlavaQwenForCausalLM"
|
| 5 |
],
|
|
@@ -173,7 +172,7 @@
|
|
| 173 |
"mm_use_im_start_end": false,
|
| 174 |
"mm_vision_select_feature": "patch",
|
| 175 |
"mm_vision_select_layer": -2,
|
| 176 |
-
"mm_vision_tower": "/
|
| 177 |
"mm_vision_tower_lr": 2e-06,
|
| 178 |
"model_type": "qwen2",
|
| 179 |
"num_attention_heads": 12,
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"LlavaQwenForCausalLM"
|
| 4 |
],
|
|
|
|
| 172 |
"mm_use_im_start_end": false,
|
| 173 |
"mm_vision_select_feature": "patch",
|
| 174 |
"mm_vision_select_layer": -2,
|
| 175 |
+
"mm_vision_tower": "google/siglip-so400m-patch14-384",
|
| 176 |
"mm_vision_tower_lr": 2e-06,
|
| 177 |
"model_type": "qwen2",
|
| 178 |
"num_attention_heads": 12,
|