Add model card and metadata

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,30 @@
+---
+pipeline_tag: robotics
+library_name: transformers
+license: apache-2.0
+tags: [robotics, agent, computer-vision, llm]
+---
+
+# <span style="font-size:30px;">STEVE-R1: Towards Long Reasoning Computer-use Agents</span>
+
+[![Hugging Face Paper](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Paper-orange)](https://huggingface.co/papers/2503.12532)
+[![Hugging Face Models](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Models-blue)](https://huggingface.co/Fanbin/STEVE-R1-7B-SFT)
+[![Hugging Face Data](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Data-green)](https://huggingface.co/datasets/Fanbin/waa_steve_trajectories)
+
+
+We evaluate the performance of the **STEVE-R1 agent** on both in-domain WindowsAgentArena (Windows 11 OS) and out-of-domain OSWorld (Ubuntu OS) benchmarks. The evaluation involves 16 attempts per task, with task completion rates recorded as the primary metric. In the in-domain Windows 11 setting, the STEVE-R1 agent demonstrated a **14%** higher task completion rate compared to the previous open-source state-of-the-art model, UI-TARS-7B-DPO. Furthermore, in the out-of-domain Ubuntu OS environment, where STEVE-R1 was not explicitly trained, it still achieved a **7%** higher task completion rate than UI-TARS-7B-DPO.
+
+<div align=center>
+<img width="98%" src="assets/performance.png"/>
+</div>
+
+## Release
+- Currently only the SFT STEVE-R1 model with step-verified training data is released. RL tunning is in progress.
+- 🔥 An improved version **STEVE-R1** is released with long reasoning ability and long image context. We extend the model context length to <b>128K</b> with at most <b>32 screenshot</b> inputs for a single task. The model response length is greatly improved with deepseek-R1 distillation, see the [examples](https://github.com/FanbinLu/STEVE-R1/tree/main/examples). We release the [training data](), [models](https://huggingface.co/Fanbin/STEVE-R1-7B-SFT), and [evaluation trajectories](https://huggingface.co/datasets/Fanbin/waa_steve_trajectories).
+- We release the paper of STEVE: Step Verification Pipeline for Computer-use Agent Training. We propose a single-frame computer-use 7B agent trained with SFT & step-verified KTO. 
+
+
+## Trajectory Data
+... (rest of the original README content)
+## Citation
+To be added.

added_tokens.json

@@ -0,0 +1,16 @@
+{
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "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 %}"
+}

config.json

@@ -0,0 +1,78 @@
+{
+  "_name_or_path": "Steve_uitars_7B_128k_sft0218_fullcorrect_with_single_frame_hf/",
+  "architectures": [
+    "Qwen2VLModel"
+  ],
+  "asr_align": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "check_data_modality": false,
+  "compress": false,
+  "compress_gap": 33,
+  "down_factor_1": 1,
+  "down_factor_2": 1,
+  "eos_token_id": 151645,
+  "freeze_mm_mlp_adapter": true,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "image_aspect_ratio": "qwen2vl",
+  "image_global": false,
+  "image_grid": 1,
+  "image_grid_pinpoints": null,
+  "image_token_id": 151655,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "keep_global": false,
+  "keep_rate": 1.0,
+  "max_position_embeddings": 128000,
+  "max_window_layers": 28,
+  "mm_hidden_size": 3584,
+  "mm_projector_lr": null,
+  "mm_projector_type": "identity",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_use_mrope": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -1,
+  "mm_vision_tower": "model_zoo/vision/UI-TARS-7B-DPO-ViT-Qwen2VL",
+  "model_type": "qwen2_vl",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "overlap_ratio": 0.0,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": {
+    "mrope_section": [
+      16,
+      24,
+      24
+    ],
+    "rope_type": "default",
+    "type": "default"
+  },
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "speech_overlap": false,
+  "speech_overlap_time": 30,
+  "tie_word_embeddings": false,
+  "tokenizer_model_max_length": 128000,
+  "tokenizer_padding_side": "right",
+  "torch_dtype": "float32",
+  "train_modality": "text_image",
+  "transformers_version": "4.49.0",
+  "tune_mm_mlp_adapter": false,
+  "use_cache": true,
+  "use_mm_proj": true,
+  "use_sliding_window": false,
+  "video_token_id": 151656,
+  "vision_config": {
+    "in_chans": 3,
+    "model_type": "qwen2_vl",
+    "spatial_patch_size": 14,
+    "torch_dtype": "float32"
+  },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652,
+  "vision_token_id": 151654,
+  "vocab_size": 152064
+}

model-00001-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9965729a270ca03179081917b0387b1422d3191dc15589d82aa0e5d8d490e344
+size 4976686992

model-00002-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:734bb4db9c16b378aac441dd29d998e87642f854b7b3562d3d4d033f9c69b2a4
+size 4778621952

model-00003-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:726f9fd34849de8352261deed0db286ce07d0c2e688ed6fc61fee48d2da7b9e5
+size 4932743600

model-00004-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbcaf2ef31145ed34c934e69a9dc8a5d4c25aa6dd1be3a6389d675705f38494e
+size 4932743624

model-00005-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fac33bea4278b321b091df1a5fccac8254713a8a272f8e99948c9022a4e55817
+size 4998851880

model-00006-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c040f796676d16ff89cc663a1dd6c754b14910070543850f451649715a192b3
+size 3662864928

model.safetensors.index.json

@@ -0,0 +1,345 @@
+{
+  "metadata": {
+    "total_size": 28282476544
+  },
+  "weight_map": {
+    "embed_tokens.weight": "model-00001-of-00006.safetensors",
+    "layers.0.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "layers.0.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.0.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.0.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.0.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "layers.0.self_attn.k_proj.bias": "model-00001-of-00006.safetensors",
+    "layers.0.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.0.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.0.self_attn.q_proj.bias": "model-00001-of-00006.safetensors",
+    "layers.0.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.0.self_attn.v_proj.bias": "model-00001-of-00006.safetensors",
+    "layers.0.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.1.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "layers.1.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.1.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.1.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.1.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "layers.1.self_attn.k_proj.bias": "model-00001-of-00006.safetensors",
+    "layers.1.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.1.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.1.self_attn.q_proj.bias": "model-00001-of-00006.safetensors",
+    "layers.1.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.1.self_attn.v_proj.bias": "model-00001-of-00006.safetensors",
+    "layers.1.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.10.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "layers.10.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.10.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.10.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.10.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "layers.10.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.10.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.10.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.10.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.10.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.10.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.10.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.11.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "layers.11.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.11.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.11.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.11.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "layers.11.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.11.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.11.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.11.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.11.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.11.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.11.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.12.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "layers.12.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.12.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.12.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.12.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "layers.12.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.12.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.12.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.12.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.12.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.12.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.12.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.13.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "layers.13.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.13.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.13.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.13.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "layers.13.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.13.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.13.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.13.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.13.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.13.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.13.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.14.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "layers.14.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.14.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.14.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.14.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "layers.14.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.14.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.14.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.14.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.14.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.14.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.14.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.15.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "layers.15.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.15.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.15.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.15.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "layers.15.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.15.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.15.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.15.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.15.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.15.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.15.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.16.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "layers.16.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.16.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.16.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.16.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "layers.16.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.16.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.16.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.16.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.16.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.16.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.16.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.17.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "layers.17.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.17.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.17.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.17.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "layers.17.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.17.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.17.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.17.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.17.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.17.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.17.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.18.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.18.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.18.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.18.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.18.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.18.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.18.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.18.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.18.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.18.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.18.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "layers.18.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "layers.19.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.19.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.19.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.19.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.19.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.19.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.19.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.19.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.19.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.19.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.19.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.19.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.2.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "layers.2.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.2.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.2.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.2.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "layers.2.self_attn.k_proj.bias": "model-00001-of-00006.safetensors",
+    "layers.2.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.2.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.2.self_attn.q_proj.bias": "model-00001-of-00006.safetensors",
+    "layers.2.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.2.self_attn.v_proj.bias": "model-00001-of-00006.safetensors",
+    "layers.2.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "layers.20.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.20.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.20.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.20.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.20.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.20.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.20.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.20.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.20.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.20.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.20.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.20.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.21.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.21.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.21.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.21.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.21.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.21.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.21.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.21.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.21.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.21.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.21.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.21.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.22.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.22.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.22.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.22.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.22.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.22.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.22.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.22.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.22.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.22.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.22.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.22.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.23.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.23.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.23.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.23.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.23.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "layers.23.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.23.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.23.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.23.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.23.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.23.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.23.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.24.input_layernorm.weight": "model-00006-of-00006.safetensors",
+    "layers.24.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.24.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.24.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.24.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "layers.24.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.24.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.24.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.24.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.24.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.24.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "layers.24.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "layers.25.input_layernorm.weight": "model-00006-of-00006.safetensors",
+    "layers.25.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.25.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.25.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.25.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "layers.25.self_attn.k_proj.bias": "model-00006-of-00006.safetensors",
+    "layers.25.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.25.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.25.self_attn.q_proj.bias": "model-00006-of-00006.safetensors",
+    "layers.25.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.25.self_attn.v_proj.bias": "model-00006-of-00006.safetensors",
+    "layers.25.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.26.input_layernorm.weight": "model-00006-of-00006.safetensors",
+    "layers.26.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.26.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.26.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.26.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "layers.26.self_attn.k_proj.bias": "model-00006-of-00006.safetensors",
+    "layers.26.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.26.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.26.self_attn.q_proj.bias": "model-00006-of-00006.safetensors",
+    "layers.26.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.26.self_attn.v_proj.bias": "model-00006-of-00006.safetensors",
+    "layers.26.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.27.input_layernorm.weight": "model-00006-of-00006.safetensors",
+    "layers.27.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.27.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.27.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.27.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "layers.27.self_attn.k_proj.bias": "model-00006-of-00006.safetensors",
+    "layers.27.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.27.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.27.self_attn.q_proj.bias": "model-00006-of-00006.safetensors",
+    "layers.27.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.27.self_attn.v_proj.bias": "model-00006-of-00006.safetensors",
+    "layers.27.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
+    "layers.3.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "layers.3.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.3.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.3.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.3.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "layers.3.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.3.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.3.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.3.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.3.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.3.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.3.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.4.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "layers.4.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.4.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.4.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.4.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "layers.4.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.4.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.4.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.4.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.4.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.4.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.4.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.5.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "layers.5.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.5.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.5.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.5.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "layers.5.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.5.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.5.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.5.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.5.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.5.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.5.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.6.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "layers.6.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.6.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.6.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.6.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "layers.6.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.6.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.6.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.6.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.6.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.6.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.6.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.7.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "layers.7.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.7.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.7.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.7.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "layers.7.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.7.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.7.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.7.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.7.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.7.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.7.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.8.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "layers.8.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.8.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.8.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.8.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "layers.8.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.8.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.8.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.8.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.8.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.8.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "layers.8.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "layers.9.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "layers.9.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.9.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.9.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.9.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "layers.9.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.9.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.9.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.9.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.9.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "layers.9.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "layers.9.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "norm.weight": "model-00006-of-00006.safetensors"
+  }
+}

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 2116800,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "Qwen2VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 2116800,
+    "shortest_edge": 3136
+  },
+  "temporal_patch_size": 2
+}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091aa7594dc2fcfbfa06b9e3c22a5f0562ac14f30375c13af7309407a0e67b8a
+size 11420371

tokenizer_config.json

@@ -0,0 +1,145 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{ message['role'].title() + ': ' + message['content'] | trim + eos_token + '\n' }}{% endfor %}{% if add_generation_prompt %}{{ 'Assistant: ' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}