Any-to-Any
Transformers
Safetensors
janus
image-text-to-text
muiltimodal
text-to-image
unified-model
Instructions to use deepseek-community/Janus-Pro-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use deepseek-community/Janus-Pro-1B with Transformers:
# Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("deepseek-community/Janus-Pro-1B") model = AutoModelForMultimodalLM.from_pretrained("deepseek-community/Janus-Pro-1B", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Image token index -> id
Browse files- config.json +1 -1
config.json
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
"architectures": [
|
| 3 |
"JanusForConditionalGeneration"
|
| 4 |
],
|
| 5 |
-
"
|
| 6 |
"model_type": "janus",
|
| 7 |
"text_config": {
|
| 8 |
"attention_bias": false,
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"JanusForConditionalGeneration"
|
| 4 |
],
|
| 5 |
+
"image_token_id": 100581,
|
| 6 |
"model_type": "janus",
|
| 7 |
"text_config": {
|
| 8 |
"attention_bias": false,
|