Add files using upload-large-folder tool

config.json

@@ -1,10 +1,9 @@
 {
-  "_name_or_path": "Qwen/Qwen2-VL-72B-Instruct",
+  "_name_or_path": "unsloth/Qwen2-VL-72B-Instruct",
   "architectures": [
     "Qwen2VLForConditionalGeneration"
   ],
   "attention_dropout": 0.0,
-  "bos_token_id": 151643,
   "eos_token_id": 151645,
   "hidden_act": "silu",
   "hidden_size": 8192,
@@ -27,7 +26,12 @@
     "bnb_4bit_use_double_quant": true,
     "llm_int8_enable_fp32_cpu_offload": false,
     "llm_int8_has_fp16_weight": false,
-    "llm_int8_skip_modules": null,
+    "llm_int8_skip_modules": [
+      "lm_head",
+      "multi_modal_projector",
+      "merger",
+      "modality_projection"
+    ],
     "llm_int8_threshold": 6.0,
     "load_in_4bit": true,
     "load_in_8bit": false,
@@ -47,7 +51,7 @@
   "sliding_window": 32768,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.46.3",
+  "transformers_version": "4.49.0",
   "unsloth_fixed": true,
   "use_cache": true,
   "use_sliding_window": false,
@@ -56,7 +60,8 @@
     "hidden_size": 8192,
     "in_chans": 3,
     "model_type": "qwen2_vl",
-    "spatial_patch_size": 14
+    "spatial_patch_size": 14,
+    "torch_dtype": "bfloat16"
   },
   "vision_end_token_id": 151653,
   "vision_start_token_id": 151652,

generation_config.json

@@ -10,5 +10,5 @@
   "repetition_penalty": 1.05,
   "top_k": 1,
   "top_p": 0.001,
-  "transformers_version": "4.46.3"
+  "transformers_version": "4.49.0"
 }

model-00001-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb657cd6fd71c36f9e32dccc9fbf63ce4edcdc91ce6c59b905b388866c2a81f4
-size 4994653119
+oid sha256:2e13715a3a74d8069d3ce48f2707a015b4d2961ad5e72e99fdd2631de1d97153
+size 4970847674

model-00002-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a3e1ad376a674af43cea78e6bef3ca47ec18ff094e4a0899a160a2204cbf75e
-size 4981068110
+oid sha256:c549de5db4f49f14ca490d8eb55dc5d6cdbbd18401ad5534a042079e20e04b62
+size 4981068103

model-00003-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:099a221caefd20edde061677478df495db15bfd1d105b40c90053cbb261765f3
-size 4981068356
+oid sha256:aa82dbbd5a0da35c3d52f55574be811561859f861ec59d9457231a15eecf0803
+size 4981068373

model-00004-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2c7918bb36f45562fa3661bbbc9b27fc9ade4b075bd0910b6bb2cac1de9a088
-size 4981068368
+oid sha256:65bdd4c5d14745baacb47035550b14e5b91bc8e916d8f24f892796abc2e63db6
+size 4981068384

model-00005-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3fee0a6a61d2069170a6d05e9af28db5d8aedda19d2c1a80eaa4ab0fb1e134a0
-size 4981068366
+oid sha256:d25758df29ad0229f4c511766fcbfa0a1c4fd61c8363483b808aa82e76a53314
+size 4981068383

model-00006-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53df35a4aaecd9229b86e2d1bc264c134de541099a7f7a3c73a0bf6c4d6c8cbb
-size 4981068358
+oid sha256:e9a9a17004df3311144e7b8f739cf43d20c4ed784c2698735d42eab662885ef4
+size 4981068372

model-00007-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:007083b98c30c543981785b022d4ca8a2f0a7f75c690925059c8f2155e29059a
-size 4981068353
+oid sha256:d3f93c26d32b06d3de4cfb76d48924b88c8c29204a7bd6e73ccdac99e486f3b8
+size 4981068371

model-00008-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3c51503da3bd8d9b09d38fbeef232de7179aec7ee86b20b5d891c3a33e5ece2
-size 4200425336
+oid sha256:1779cff4999f4c72280f47ef1c40f421913d8b6f6b9dd1821c5efbf5f90e5f8f
+size 4325381691

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_size": 41572342422
+    "total_size": 41673494463
   },
   "weight_map": {
     "lm_head.weight": "model-00009-of-00009.safetensors",
@@ -347,12 +347,12 @@
     "model.layers.15.mlp.gate_proj.weight.nested_quant_map": "model-00002-of-00009.safetensors",
     "model.layers.15.mlp.gate_proj.weight.quant_map": "model-00002-of-00009.safetensors",
     "model.layers.15.mlp.gate_proj.weight.quant_state.bitsandbytes__nf4": "model-00002-of-00009.safetensors",
-    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00009.safetensors",
-    "model.layers.15.mlp.up_proj.weight.absmax": "model-00002-of-00009.safetensors",
-    "model.layers.15.mlp.up_proj.weight.nested_absmax": "model-00002-of-00009.safetensors",
-    "model.layers.15.mlp.up_proj.weight.nested_quant_map": "model-00002-of-00009.safetensors",
-    "model.layers.15.mlp.up_proj.weight.quant_map": "model-00002-of-00009.safetensors",
-    "model.layers.15.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00002-of-00009.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00003-of-00009.safetensors",
+    "model.layers.15.mlp.up_proj.weight.absmax": "model-00003-of-00009.safetensors",
+    "model.layers.15.mlp.up_proj.weight.nested_absmax": "model-00003-of-00009.safetensors",
+    "model.layers.15.mlp.up_proj.weight.nested_quant_map": "model-00003-of-00009.safetensors",
+    "model.layers.15.mlp.up_proj.weight.quant_map": "model-00003-of-00009.safetensors",
+    "model.layers.15.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00003-of-00009.safetensors",
     "model.layers.15.post_attention_layernorm.weight": "model-00003-of-00009.safetensors",
     "model.layers.15.self_attn.k_proj.bias": "model-00002-of-00009.safetensors",
     "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00009.safetensors",
@@ -911,12 +911,12 @@
     "model.layers.26.mlp.gate_proj.weight.nested_quant_map": "model-00003-of-00009.safetensors",
     "model.layers.26.mlp.gate_proj.weight.quant_map": "model-00003-of-00009.safetensors",
     "model.layers.26.mlp.gate_proj.weight.quant_state.bitsandbytes__nf4": "model-00003-of-00009.safetensors",
-    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00009.safetensors",
-    "model.layers.26.mlp.up_proj.weight.absmax": "model-00003-of-00009.safetensors",
-    "model.layers.26.mlp.up_proj.weight.nested_absmax": "model-00003-of-00009.safetensors",
-    "model.layers.26.mlp.up_proj.weight.nested_quant_map": "model-00003-of-00009.safetensors",
-    "model.layers.26.mlp.up_proj.weight.quant_map": "model-00003-of-00009.safetensors",
-    "model.layers.26.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00003-of-00009.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00004-of-00009.safetensors",
+    "model.layers.26.mlp.up_proj.weight.absmax": "model-00004-of-00009.safetensors",
+    "model.layers.26.mlp.up_proj.weight.nested_absmax": "model-00004-of-00009.safetensors",
+    "model.layers.26.mlp.up_proj.weight.nested_quant_map": "model-00004-of-00009.safetensors",
+    "model.layers.26.mlp.up_proj.weight.quant_map": "model-00004-of-00009.safetensors",
+    "model.layers.26.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00004-of-00009.safetensors",
     "model.layers.26.post_attention_layernorm.weight": "model-00004-of-00009.safetensors",
     "model.layers.26.self_attn.k_proj.bias": "model-00003-of-00009.safetensors",
     "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00009.safetensors",
@@ -1475,12 +1475,12 @@
     "model.layers.37.mlp.gate_proj.weight.nested_quant_map": "model-00004-of-00009.safetensors",
     "model.layers.37.mlp.gate_proj.weight.quant_map": "model-00004-of-00009.safetensors",
     "model.layers.37.mlp.gate_proj.weight.quant_state.bitsandbytes__nf4": "model-00004-of-00009.safetensors",
-    "model.layers.37.mlp.up_proj.weight": "model-00004-of-00009.safetensors",
-    "model.layers.37.mlp.up_proj.weight.absmax": "model-00004-of-00009.safetensors",
-    "model.layers.37.mlp.up_proj.weight.nested_absmax": "model-00004-of-00009.safetensors",
-    "model.layers.37.mlp.up_proj.weight.nested_quant_map": "model-00004-of-00009.safetensors",
-    "model.layers.37.mlp.up_proj.weight.quant_map": "model-00004-of-00009.safetensors",
-    "model.layers.37.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00004-of-00009.safetensors",
+    "model.layers.37.mlp.up_proj.weight": "model-00005-of-00009.safetensors",
+    "model.layers.37.mlp.up_proj.weight.absmax": "model-00005-of-00009.safetensors",
+    "model.layers.37.mlp.up_proj.weight.nested_absmax": "model-00005-of-00009.safetensors",
+    "model.layers.37.mlp.up_proj.weight.nested_quant_map": "model-00005-of-00009.safetensors",
+    "model.layers.37.mlp.up_proj.weight.quant_map": "model-00005-of-00009.safetensors",
+    "model.layers.37.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00005-of-00009.safetensors",
     "model.layers.37.post_attention_layernorm.weight": "model-00005-of-00009.safetensors",
     "model.layers.37.self_attn.k_proj.bias": "model-00004-of-00009.safetensors",
     "model.layers.37.self_attn.k_proj.weight": "model-00004-of-00009.safetensors",
@@ -1616,12 +1616,12 @@
     "model.layers.4.mlp.gate_proj.weight.nested_quant_map": "model-00001-of-00009.safetensors",
     "model.layers.4.mlp.gate_proj.weight.quant_map": "model-00001-of-00009.safetensors",
     "model.layers.4.mlp.gate_proj.weight.quant_state.bitsandbytes__nf4": "model-00001-of-00009.safetensors",
-    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00009.safetensors",
-    "model.layers.4.mlp.up_proj.weight.absmax": "model-00001-of-00009.safetensors",
-    "model.layers.4.mlp.up_proj.weight.nested_absmax": "model-00001-of-00009.safetensors",
-    "model.layers.4.mlp.up_proj.weight.nested_quant_map": "model-00001-of-00009.safetensors",
-    "model.layers.4.mlp.up_proj.weight.quant_map": "model-00001-of-00009.safetensors",
-    "model.layers.4.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00001-of-00009.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00002-of-00009.safetensors",
+    "model.layers.4.mlp.up_proj.weight.absmax": "model-00002-of-00009.safetensors",
+    "model.layers.4.mlp.up_proj.weight.nested_absmax": "model-00002-of-00009.safetensors",
+    "model.layers.4.mlp.up_proj.weight.nested_quant_map": "model-00002-of-00009.safetensors",
+    "model.layers.4.mlp.up_proj.weight.quant_map": "model-00002-of-00009.safetensors",
+    "model.layers.4.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00002-of-00009.safetensors",
     "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00009.safetensors",
     "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00009.safetensors",
     "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00009.safetensors",
@@ -2039,12 +2039,12 @@
     "model.layers.48.mlp.gate_proj.weight.nested_quant_map": "model-00005-of-00009.safetensors",
     "model.layers.48.mlp.gate_proj.weight.quant_map": "model-00005-of-00009.safetensors",
     "model.layers.48.mlp.gate_proj.weight.quant_state.bitsandbytes__nf4": "model-00005-of-00009.safetensors",
-    "model.layers.48.mlp.up_proj.weight": "model-00005-of-00009.safetensors",
-    "model.layers.48.mlp.up_proj.weight.absmax": "model-00005-of-00009.safetensors",
-    "model.layers.48.mlp.up_proj.weight.nested_absmax": "model-00005-of-00009.safetensors",
-    "model.layers.48.mlp.up_proj.weight.nested_quant_map": "model-00005-of-00009.safetensors",
-    "model.layers.48.mlp.up_proj.weight.quant_map": "model-00005-of-00009.safetensors",
-    "model.layers.48.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00005-of-00009.safetensors",
+    "model.layers.48.mlp.up_proj.weight": "model-00006-of-00009.safetensors",
+    "model.layers.48.mlp.up_proj.weight.absmax": "model-00006-of-00009.safetensors",
+    "model.layers.48.mlp.up_proj.weight.nested_absmax": "model-00006-of-00009.safetensors",
+    "model.layers.48.mlp.up_proj.weight.nested_quant_map": "model-00006-of-00009.safetensors",
+    "model.layers.48.mlp.up_proj.weight.quant_map": "model-00006-of-00009.safetensors",
+    "model.layers.48.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00006-of-00009.safetensors",
     "model.layers.48.post_attention_layernorm.weight": "model-00006-of-00009.safetensors",
     "model.layers.48.self_attn.k_proj.bias": "model-00005-of-00009.safetensors",
     "model.layers.48.self_attn.k_proj.weight": "model-00005-of-00009.safetensors",
@@ -2603,12 +2603,12 @@
     "model.layers.59.mlp.gate_proj.weight.nested_quant_map": "model-00006-of-00009.safetensors",
     "model.layers.59.mlp.gate_proj.weight.quant_map": "model-00006-of-00009.safetensors",
     "model.layers.59.mlp.gate_proj.weight.quant_state.bitsandbytes__nf4": "model-00006-of-00009.safetensors",
-    "model.layers.59.mlp.up_proj.weight": "model-00006-of-00009.safetensors",
-    "model.layers.59.mlp.up_proj.weight.absmax": "model-00006-of-00009.safetensors",
-    "model.layers.59.mlp.up_proj.weight.nested_absmax": "model-00006-of-00009.safetensors",
-    "model.layers.59.mlp.up_proj.weight.nested_quant_map": "model-00006-of-00009.safetensors",
-    "model.layers.59.mlp.up_proj.weight.quant_map": "model-00006-of-00009.safetensors",
-    "model.layers.59.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00006-of-00009.safetensors",
+    "model.layers.59.mlp.up_proj.weight": "model-00007-of-00009.safetensors",
+    "model.layers.59.mlp.up_proj.weight.absmax": "model-00007-of-00009.safetensors",
+    "model.layers.59.mlp.up_proj.weight.nested_absmax": "model-00007-of-00009.safetensors",
+    "model.layers.59.mlp.up_proj.weight.nested_quant_map": "model-00007-of-00009.safetensors",
+    "model.layers.59.mlp.up_proj.weight.quant_map": "model-00007-of-00009.safetensors",
+    "model.layers.59.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00007-of-00009.safetensors",
     "model.layers.59.post_attention_layernorm.weight": "model-00007-of-00009.safetensors",
     "model.layers.59.self_attn.k_proj.bias": "model-00006-of-00009.safetensors",
     "model.layers.59.self_attn.k_proj.weight": "model-00006-of-00009.safetensors",
@@ -3214,12 +3214,12 @@
     "model.layers.70.mlp.gate_proj.weight.nested_quant_map": "model-00007-of-00009.safetensors",
     "model.layers.70.mlp.gate_proj.weight.quant_map": "model-00007-of-00009.safetensors",
     "model.layers.70.mlp.gate_proj.weight.quant_state.bitsandbytes__nf4": "model-00007-of-00009.safetensors",
-    "model.layers.70.mlp.up_proj.weight": "model-00007-of-00009.safetensors",
-    "model.layers.70.mlp.up_proj.weight.absmax": "model-00007-of-00009.safetensors",
-    "model.layers.70.mlp.up_proj.weight.nested_absmax": "model-00007-of-00009.safetensors",
-    "model.layers.70.mlp.up_proj.weight.nested_quant_map": "model-00007-of-00009.safetensors",
-    "model.layers.70.mlp.up_proj.weight.quant_map": "model-00007-of-00009.safetensors",
-    "model.layers.70.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00007-of-00009.safetensors",
+    "model.layers.70.mlp.up_proj.weight": "model-00008-of-00009.safetensors",
+    "model.layers.70.mlp.up_proj.weight.absmax": "model-00008-of-00009.safetensors",
+    "model.layers.70.mlp.up_proj.weight.nested_absmax": "model-00008-of-00009.safetensors",
+    "model.layers.70.mlp.up_proj.weight.nested_quant_map": "model-00008-of-00009.safetensors",
+    "model.layers.70.mlp.up_proj.weight.quant_map": "model-00008-of-00009.safetensors",
+    "model.layers.70.mlp.up_proj.weight.quant_state.bitsandbytes__nf4": "model-00008-of-00009.safetensors",
     "model.layers.70.post_attention_layernorm.weight": "model-00008-of-00009.safetensors",
     "model.layers.70.self_attn.k_proj.bias": "model-00007-of-00009.safetensors",
     "model.layers.70.self_attn.k_proj.weight": "model-00007-of-00009.safetensors",
@@ -4794,18 +4794,8 @@
     "visual.merger.ln_q.weight": "model-00001-of-00009.safetensors",
     "visual.merger.mlp.0.bias": "model-00001-of-00009.safetensors",
     "visual.merger.mlp.0.weight": "model-00001-of-00009.safetensors",
-    "visual.merger.mlp.0.weight.absmax": "model-00001-of-00009.safetensors",
-    "visual.merger.mlp.0.weight.nested_absmax": "model-00001-of-00009.safetensors",
-    "visual.merger.mlp.0.weight.nested_quant_map": "model-00001-of-00009.safetensors",
-    "visual.merger.mlp.0.weight.quant_map": "model-00001-of-00009.safetensors",
-    "visual.merger.mlp.0.weight.quant_state.bitsandbytes__nf4": "model-00001-of-00009.safetensors",
     "visual.merger.mlp.2.bias": "model-00001-of-00009.safetensors",
     "visual.merger.mlp.2.weight": "model-00001-of-00009.safetensors",
-    "visual.merger.mlp.2.weight.absmax": "model-00001-of-00009.safetensors",
-    "visual.merger.mlp.2.weight.nested_absmax": "model-00001-of-00009.safetensors",
-    "visual.merger.mlp.2.weight.nested_quant_map": "model-00001-of-00009.safetensors",
-    "visual.merger.mlp.2.weight.quant_map": "model-00001-of-00009.safetensors",
-    "visual.merger.mlp.2.weight.quant_state.bitsandbytes__nf4": "model-00001-of-00009.safetensors",
     "visual.patch_embed.proj.weight": "model-00001-of-00009.safetensors"
   }
 }

preprocessor_config.json

@@ -22,8 +22,8 @@
   "resample": 3,
   "rescale_factor": 0.00392156862745098,
   "size": {
-    "max_pixels": 12845056,
-    "min_pixels": 3136
+    "longest_edge": 12845056,
+    "shortest_edge": 3136
   },
   "temporal_patch_size": 2
 }

tokenizer_config.json

@@ -134,6 +134,7 @@
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|im_end|>",
   "errors": "replace",
+  "extra_special_tokens": {},
   "model_max_length": 32768,
   "pad_token": "<|vision_pad|>",
   "padding_side": "left",