Instructions to use tiny-random/minicpm-v-4.6 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use tiny-random/minicpm-v-4.6 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="tiny-random/minicpm-v-4.6")# Load model directly from transformers import AutoProcessor, AutoModel processor = AutoProcessor.from_pretrained("tiny-random/minicpm-v-4.6") model = AutoModel.from_pretrained("tiny-random/minicpm-v-4.6") - Notebooks
- Google Colab
- Kaggle
| { | |
| "architectures": [ | |
| "MiniCPMV4_6Model" | |
| ], | |
| "bos_token_id": null, | |
| "downsample_mode": "16x", | |
| "drop_vision_last_layer": false, | |
| "dtype": "bfloat16", | |
| "eos_token_id": 248044, | |
| "image_size": 1120, | |
| "image_token_id": 248056, | |
| "insert_layer_id": 6, | |
| "merge_kernel_size": [ | |
| 2, | |
| 2 | |
| ], | |
| "merger_times": 1, | |
| "model_type": "minicpmv4_6", | |
| "pad_token_id": null, | |
| "patch_size": 14, | |
| "text_config": { | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "attn_output_gate": true, | |
| "bos_token_id": null, | |
| "eos_token_id": null, | |
| "full_attention_interval": 4, | |
| "head_dim": 32, | |
| "hidden_act": "silu", | |
| "hidden_size": 8, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 64, | |
| "layer_types": [ | |
| "linear_attention", | |
| "linear_attention", | |
| "linear_attention", | |
| "full_attention" | |
| ], | |
| "linear_conv_kernel_dim": 4, | |
| "linear_key_head_dim": 32, | |
| "linear_num_key_heads": 4, | |
| "linear_num_value_heads": 4, | |
| "linear_value_head_dim": 32, | |
| "mamba_ssm_dtype": "float32", | |
| "max_position_embeddings": 262144, | |
| "mlp_only_layers": [], | |
| "model_type": "qwen3_5_text", | |
| "mtp_num_hidden_layers": 1, | |
| "mtp_use_dedicated_embeddings": false, | |
| "num_attention_heads": 8, | |
| "num_hidden_layers": 4, | |
| "num_key_value_heads": 2, | |
| "pad_token_id": null, | |
| "partial_rotary_factor": 0.25, | |
| "rms_norm_eps": 1e-06, | |
| "rope_parameters": { | |
| "partial_rotary_factor": 0.25, | |
| "rope_theta": 10000000, | |
| "rope_type": "default" | |
| }, | |
| "tie_word_embeddings": true, | |
| "use_cache": true, | |
| "vocab_size": 248094 | |
| }, | |
| "tie_word_embeddings": true, | |
| "transformers_version": "5.9.0", | |
| "video_token_id": 248057, | |
| "vision_config": { | |
| "attention_dropout": 0.0, | |
| "hidden_act": "gelu_pytorch_tanh", | |
| "hidden_size": 128, | |
| "image_size": 980, | |
| "insert_layer_id": 6, | |
| "intermediate_size": 128, | |
| "layer_norm_eps": 1e-06, | |
| "model_type": "minicpmv4_6_vision", | |
| "num_attention_heads": 4, | |
| "num_channels": 3, | |
| "num_hidden_layers": 2, | |
| "patch_size": 14, | |
| "window_kernel_size": [ | |
| 2, | |
| 2 | |
| ] | |
| } | |
| } | |