Transformers
Safetensors
vision-language
document-understanding
boundingdocs
bitsandbytes
8-bit precision
Instructions to use CompressingVLM/glm-ocr-boundingdocs-ft-bnb-int8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use CompressingVLM/glm-ocr-boundingdocs-ft-bnb-int8 with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("CompressingVLM/glm-ocr-boundingdocs-ft-bnb-int8", dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "architectures": [ | |
| "GlmOcrForConditionalGeneration" | |
| ], | |
| "dtype": "bfloat16", | |
| "image_end_token_id": 59257, | |
| "image_start_token_id": 59256, | |
| "image_token_id": 59280, | |
| "model_type": "glm_ocr", | |
| "quantization_config": { | |
| "_load_in_4bit": false, | |
| "_load_in_8bit": true, | |
| "bnb_4bit_compute_dtype": "float32", | |
| "bnb_4bit_quant_storage": "uint8", | |
| "bnb_4bit_quant_type": "fp4", | |
| "bnb_4bit_use_double_quant": false, | |
| "llm_int8_enable_fp32_cpu_offload": false, | |
| "llm_int8_has_fp16_weight": false, | |
| "llm_int8_skip_modules": null, | |
| "llm_int8_threshold": 6.0, | |
| "load_in_4bit": false, | |
| "load_in_8bit": true, | |
| "quant_method": "bitsandbytes" | |
| }, | |
| "text_config": { | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "dtype": "bfloat16", | |
| "eos_token_id": [ | |
| 59246, | |
| 59253 | |
| ], | |
| "head_dim": 128, | |
| "hidden_act": "silu", | |
| "hidden_size": 1536, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 4608, | |
| "max_position_embeddings": 131072, | |
| "model_type": "glm_ocr_text", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 16, | |
| "num_key_value_heads": 8, | |
| "num_nextn_predict_layers": 1, | |
| "pad_token_id": 59246, | |
| "rms_norm_eps": 1e-05, | |
| "rope_parameters": { | |
| "mrope_section": [ | |
| 16, | |
| 24, | |
| 24 | |
| ], | |
| "partial_rotary_factor": 1.0, | |
| "rope_theta": 10000, | |
| "rope_type": "default" | |
| }, | |
| "tie_word_embeddings": false, | |
| "use_cache": true, | |
| "vocab_size": 59392 | |
| }, | |
| "tie_word_embeddings": false, | |
| "transformers_version": "5.9.0", | |
| "video_end_token_id": 59259, | |
| "video_start_token_id": 59258, | |
| "video_token_id": 59281, | |
| "vision_config": { | |
| "attention_bias": true, | |
| "attention_dropout": 0.0, | |
| "depth": 24, | |
| "dtype": "bfloat16", | |
| "hidden_act": "silu", | |
| "hidden_dropout_prob": 0.0, | |
| "hidden_size": 1024, | |
| "image_size": 336, | |
| "in_channels": 3, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 4096, | |
| "model_type": "glm_ocr_vision", | |
| "num_heads": 16, | |
| "out_hidden_size": 1536, | |
| "patch_size": 14, | |
| "rms_norm_eps": 1e-05, | |
| "spatial_merge_size": 2, | |
| "temporal_patch_size": 2 | |
| } | |
| } | |