Upload folder using huggingface_hub

config.json

@@ -3,13 +3,14 @@
     "Qwen2_5_VLForConditionalGeneration"
   ],
   "attention_dropout": 0.0,
+  "dtype": "bfloat16",
   "eos_token_id": 151645,
   "hidden_act": "silu",
   "hidden_size": 3584,
   "image_token_id": 151655,
   "initializer_range": 0.02,
   "intermediate_size": 18944,
-  "max_position_embeddings": 128000,
+  "max_position_embeddings": 32768,
   "max_window_layers": 28,
   "model_type": "qwen2_5_vl",
   "num_attention_heads": 28,
@@ -26,72 +27,75 @@
     "llm_int8_enable_fp32_cpu_offload": false,
     "llm_int8_has_fp16_weight": false,
     "llm_int8_skip_modules": [
+      "embed_tokens",
+      "embedding",
       "lm_head",
       "multi_modal_projector",
       "merger",
       "modality_projection",
-      "visual.blocks.27.attn",
-      "visual.blocks.28.attn",
-      "visual.blocks.25.attn",
-      "visual.blocks.22.attn",
-      "visual.blocks.21.attn",
-      "visual.blocks.29.mlp",
-      "visual.blocks.24.attn",
-      "visual.blocks.29.attn",
-      "visual.blocks.31.attn",
-      "visual.blocks.30.attn",
-      "visual.blocks.28.mlp",
-      "visual.blocks.20.attn",
-      "visual.blocks.25.mlp",
-      "visual.blocks.19.attn",
-      "visual.blocks.26.mlp",
-      "visual.blocks.24.mlp",
-      "visual.blocks.17.attn",
-      "visual.blocks.27.mlp",
-      "visual.blocks.23.attn",
-      "visual.blocks.23.mlp",
-      "visual.blocks.21.mlp",
-      "visual.blocks.19.mlp",
-      "visual.blocks.18.attn",
-      "visual.blocks.20.mlp",
-      "visual.blocks.11.attn",
-      "visual.blocks.9.mlp",
-      "visual.blocks.9.attn",
-      "visual.blocks.16.attn",
-      "visual.blocks.11.mlp",
-      "visual.blocks.22.mlp",
-      "visual.blocks.18.mlp",
-      "visual.blocks.13.attn",
-      "visual.blocks.12.attn",
-      "visual.blocks.6.attn",
-      "visual.blocks.10.mlp",
-      "visual.blocks.8.mlp",
-      "visual.blocks.8.attn",
-      "visual.blocks.14.attn",
-      "visual.blocks.4.mlp",
-      "visual.blocks.16.mlp",
-      "visual.blocks.7.mlp",
-      "visual.blocks.6.mlp",
-      "visual.blocks.15.mlp",
-      "visual.blocks.5.mlp",
-      "visual.blocks.10.attn",
-      "visual.blocks.3.mlp",
-      "visual.blocks.12.mlp",
-      "visual.blocks.13.mlp",
-      "visual.blocks.14.mlp",
-      "visual.blocks.2.mlp",
-      "visual.blocks.5.attn",
-      "visual.blocks.1.attn",
-      "visual.blocks.2.attn",
-      "visual.blocks.4.attn",
-      "visual.blocks.3.attn",
-      "visual.blocks.15.attn",
-      "visual.blocks.1.mlp",
-      "visual.blocks.17.mlp",
-      "visual.blocks.0.attn",
-      "visual.blocks.7.attn",
-      "visual.blocks.0.mlp",
-      "visual.blocks.31.mlp.down_proj"
+      "router",
+      "model.visual.blocks.31.mlp",
+      "model.visual.blocks.24.attn",
+      "model.visual.blocks.30.mlp",
+      "model.visual.blocks.30.attn",
+      "model.visual.blocks.25.attn",
+      "model.visual.blocks.29.attn",
+      "model.visual.blocks.26.attn",
+      "model.visual.blocks.28.attn",
+      "model.visual.blocks.19.attn",
+      "model.visual.blocks.29.mlp",
+      "model.visual.blocks.28.mlp",
+      "model.visual.blocks.31.attn",
+      "model.visual.blocks.25.mlp",
+      "model.visual.blocks.26.mlp",
+      "model.visual.blocks.20.attn",
+      "model.visual.blocks.27.mlp",
+      "model.visual.blocks.17.attn",
+      "model.visual.blocks.24.mlp",
+      "model.visual.blocks.18.attn",
+      "model.visual.blocks.16.attn",
+      "model.visual.blocks.11.attn",
+      "model.visual.blocks.21.mlp",
+      "model.visual.blocks.20.mlp",
+      "model.visual.blocks.23.mlp",
+      "model.visual.blocks.9.attn",
+      "model.visual.blocks.12.attn",
+      "model.visual.blocks.23.attn",
+      "model.visual.blocks.19.mlp",
+      "model.visual.blocks.22.mlp",
+      "model.visual.blocks.18.mlp",
+      "model.visual.blocks.13.attn",
+      "model.visual.blocks.8.attn",
+      "model.visual.blocks.11.mlp",
+      "model.visual.blocks.10.mlp",
+      "model.visual.blocks.6.attn",
+      "model.visual.blocks.15.mlp",
+      "model.visual.blocks.8.mlp",
+      "model.visual.blocks.9.mlp",
+      "model.visual.blocks.14.attn",
+      "model.visual.blocks.5.mlp",
+      "model.visual.blocks.14.mlp",
+      "model.visual.blocks.10.attn",
+      "model.visual.blocks.6.mlp",
+      "model.visual.blocks.7.mlp",
+      "model.visual.blocks.5.attn",
+      "model.visual.blocks.4.mlp",
+      "model.visual.blocks.16.mlp",
+      "model.visual.blocks.12.mlp",
+      "model.visual.blocks.13.mlp",
+      "model.visual.blocks.2.mlp",
+      "model.visual.blocks.3.mlp",
+      "model.visual.blocks.1.attn",
+      "model.visual.blocks.0.attn",
+      "model.visual.blocks.4.attn",
+      "model.visual.blocks.2.attn",
+      "model.visual.blocks.15.attn",
+      "model.visual.blocks.3.attn",
+      "model.visual.blocks.1.mlp",
+      "model.visual.blocks.17.mlp",
+      "model.visual.blocks.0.mlp",
+      "model.visual.blocks.7.attn",
+      "model.visual.blocks.31.mlp.down_proj"
     ],
     "llm_int8_threshold": 6.0,
     "load_in_4bit": true,
@@ -110,15 +114,84 @@
   },
   "rope_theta": 1000000.0,
   "sliding_window": 32768,
+  "text_config": {
+    "architectures": [
+      "Qwen2_5_VLForConditionalGeneration"
+    ],
+    "attention_dropout": 0.0,
+    "bos_token_id": 151643,
+    "dtype": "bfloat16",
+    "eos_token_id": 151645,
+    "hidden_act": "silu",
+    "hidden_size": 3584,
+    "image_token_id": null,
+    "initializer_range": 0.02,
+    "intermediate_size": 18944,
+    "layer_types": [
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention"
+    ],
+    "max_position_embeddings": 128000,
+    "max_window_layers": 28,
+    "model_type": "qwen2_5_vl_text",
+    "num_attention_heads": 28,
+    "num_hidden_layers": 28,
+    "num_key_value_heads": 4,
+    "rms_norm_eps": 1e-06,
+    "rope_scaling": {
+      "mrope_section": [
+        16,
+        24,
+        24
+      ],
+      "rope_type": "default",
+      "type": "default"
+    },
+    "rope_theta": 1000000.0,
+    "sliding_window": null,
+    "use_cache": true,
+    "use_sliding_window": false,
+    "video_token_id": null,
+    "vision_end_token_id": 151653,
+    "vision_start_token_id": 151652,
+    "vision_token_id": 151654,
+    "vocab_size": 152064
+  },
   "tie_word_embeddings": false,
-  "torch_dtype": "bfloat16",
-  "transformers_version": "4.51.3",
+  "transformers_version": "4.56.1",
   "unsloth_fixed": true,
   "use_cache": true,
   "use_sliding_window": false,
   "video_token_id": 151656,
   "vision_config": {
     "depth": 32,
+    "dtype": "bfloat16",
     "fullatt_block_indexes": [
       7,
       15,
@@ -129,6 +202,7 @@
     "hidden_size": 1280,
     "in_channels": 3,
     "in_chans": 3,
+    "initializer_range": 0.02,
     "intermediate_size": 3420,
     "model_type": "qwen2_5_vl",
     "num_heads": 16,
@@ -138,7 +212,6 @@
     "spatial_patch_size": 14,
     "temporal_patch_size": 2,
     "tokens_per_second": 2,
-    "torch_dtype": "bfloat16",
     "window_size": 112
   },
   "vision_end_token_id": 151653,

generation_config.json

@@ -5,9 +5,9 @@
     151645,
     151643
   ],
-  "max_length": 128000,
+  "max_length": 32768,
   "pad_token_id": 151654,
   "repetition_penalty": 1.05,
   "temperature": 1e-06,
-  "transformers_version": "4.51.3"
+  "transformers_version": "4.56.1"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f2041306642ad99ee54b96a372a2603010ba678c86595a721c3100a4febf504
-size 6858199244
+oid sha256:7e6a35589c2fbb08b32aebd93b3d3da57b5716658487f92c91ae524dc8982661
+size 6871229143

preprocessor_config.json

@@ -1,4 +1,10 @@
 {
+  "crop_size": null,
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "device": null,
+  "disable_grouping": null,
+  "do_center_crop": null,
   "do_convert_rgb": true,
   "do_normalize": true,
   "do_rescale": true,
@@ -8,12 +14,13 @@
     0.4578275,
     0.40821073
   ],
-  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_processor_type": "Qwen2VLImageProcessorFast",
   "image_std": [
     0.26862954,
     0.26130258,
     0.27577711
   ],
+  "input_data_format": null,
   "max_pixels": 12845056,
   "merge_size": 2,
   "min_pixels": 3136,
@@ -21,6 +28,7 @@
   "processor_class": "Qwen2_5_VLProcessor",
   "resample": 3,
   "rescale_factor": 0.00392156862745098,
+  "return_tensors": null,
   "size": {
     "longest_edge": 12845056,
     "shortest_edge": 3136

tokenizer_config.json

@@ -195,16 +195,16 @@
     "<|video_pad|>"
   ],
   "bos_token": null,
-  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|im_end|>",
   "errors": "replace",
   "extra_special_tokens": {},
-  "model_max_length": 128000,
+  "model_max_length": 32768,
   "pad_token": "<|vision_pad|>",
   "padding_side": "left",
   "processor_class": "Qwen2_5_VLProcessor",
   "split_special_tokens": false,
   "tokenizer_class": "Qwen2Tokenizer",
-  "unk_token": null
+  "unk_token": null,
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
 }

video_preprocessor_config.json

@@ -0,0 +1,44 @@
+{
+  "crop_size": null,
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "device": null,
+  "do_center_crop": null,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_pad": null,
+  "do_rescale": true,
+  "do_resize": true,
+  "do_sample_frames": false,
+  "fps": null,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "input_data_format": null,
+  "max_frames": 768,
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_frames": 4,
+  "min_pixels": 3136,
+  "num_frames": null,
+  "patch_size": 14,
+  "processor_class": "Qwen2_5_VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "return_metadata": false,
+  "size": {
+    "longest_edge": 12845056,
+    "shortest_edge": 3136
+  },
+  "size_divisor": null,
+  "temporal_patch_size": 2,
+  "video_metadata": null,
+  "video_processor_type": "Qwen2VLVideoProcessor"
+}