BAAI
/

Image-Text-to-Text
Transformers
Safetensors
English
Chinese
multimodal
Inference Endpoints
ldwang's picture
Add files using upload-large-folder tool
661be46 verified
{
"_name_or_path": "/share/project/gushuhao/1.research/LLaVA-NeXT-main/llavaonevision-qwen1.5b-mlp2x_gelu-qwen2.5-stage1.5a1-1012-32n40g-node8",
"architectures": [
"LlavaQwenForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 151643,
"eos_token_id": 151645,
"hidden_act": "silu",
"hidden_size": 1536,
"image_aspect_ratio": "anyres_max_5",
"image_crop_resolution": null,
"image_grid_pinpoints": [
[
384,
384
],
[
384,
768
],
[
384,
1152
],
[
768,
384
],
[
768,
768
],
[
768,
1152
],
[
1152,
384
],
[
1152,
768
],
[
1152,
1152
]
],
"image_split_resolution": null,
"initializer_range": 0.02,
"intermediate_size": 8960,
"max_position_embeddings": 32768,
"max_window_layers": 21,
"mm_hidden_size": 1152,
"mm_newline_position": "one_token",
"mm_patch_merge_type": "spatial_unpad",
"mm_projector_lr": null,
"mm_projector_type": "mlp2x_gelu",
"mm_resampler_type": null,
"mm_spatial_pool_mode": "bilinear",
"mm_tunable_parts": "mm_vision_tower,mm_mlp_adapter,mm_language_model",
"mm_use_im_patch_token": false,
"mm_use_im_start_end": false,
"mm_vision_select_feature": "patch",
"mm_vision_select_layer": -2,
"mm_vision_tower": "/share/project/gushuhao/external_model/siglip-so400m-patch14-384",
"mm_vision_tower_lr": 2e-06,
"model_type": "qwen2",
"num_attention_heads": 12,
"num_hidden_layers": 28,
"num_key_value_heads": 2,
"pos_skipping_range": 4096,
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 1000000.0,
"sliding_window": 32768,
"tie_word_embeddings": true,
"tokenizer_model_max_length": 12288,
"tokenizer_padding_side": "right",
"torch_dtype": "bfloat16",
"transformers_version": "4.40.0",
"use_cache": true,
"use_mm_proj": true,
"use_pos_skipping": false,
"use_sliding_window": false,
"vision_tower_pretrained": null,
"vocab_size": 151936
}