Instructions to use openbmb/MiniCPM-V with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use openbmb/MiniCPM-V with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("visual-question-answering", model="openbmb/MiniCPM-V", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("openbmb/MiniCPM-V", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
fix bug
#5
by baymax591 - opened
- config.json +2 -0
config.json
CHANGED
|
@@ -24,6 +24,7 @@
|
|
| 24 |
"num_attention_heads": 36,
|
| 25 |
"num_hidden_layers": 40,
|
| 26 |
"num_key_value_heads": 36,
|
|
|
|
| 27 |
"pretraining_tp": 1,
|
| 28 |
"query_num": 64,
|
| 29 |
"rms_norm_eps": 1e-05,
|
|
@@ -31,6 +32,7 @@
|
|
| 31 |
"rope_theta": 10000.0,
|
| 32 |
"scale_depth": 1.4,
|
| 33 |
"scale_emb": 12,
|
|
|
|
| 34 |
"tie_word_embeddings": false,
|
| 35 |
"torch_dtype": "bfloat16",
|
| 36 |
"transformers_version": "4.36.0",
|
|
|
|
| 24 |
"num_attention_heads": 36,
|
| 25 |
"num_hidden_layers": 40,
|
| 26 |
"num_key_value_heads": 36,
|
| 27 |
+
"patch_size": 14,
|
| 28 |
"pretraining_tp": 1,
|
| 29 |
"query_num": 64,
|
| 30 |
"rms_norm_eps": 1e-05,
|
|
|
|
| 32 |
"rope_theta": 10000.0,
|
| 33 |
"scale_depth": 1.4,
|
| 34 |
"scale_emb": 12,
|
| 35 |
+
"scale_resolution": 448,
|
| 36 |
"tie_word_embeddings": false,
|
| 37 |
"torch_dtype": "bfloat16",
|
| 38 |
"transformers_version": "4.36.0",
|