初始化项目，由ModelHub XC社区提供模型

Model: RedHatAI/Qwen2.5-VL-3B-Instruct-quantized.w4a16 Source: Original Platform
2026-10-06 17:32:13 +08:00
commit f9e5ed2c1d
15 changed files with 984 additions and 0 deletions
--- a/.gitattributes
+++ b/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz 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
--- a/README.md
+++ b/README.md
@@ -0,0 +1,575 @@
+---
+tags:
+- vllm
+- vision
+- w4a16
+license: apache-2.0
+license_link: >-
+  https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/apache-2.0.md
+language:
+- en
+base_model: Qwen/Qwen2.5-VL-3B-Instruct
+library_name: transformers
+---
+
+# Qwen2.5-VL-3B-Instruct-quantized-w4a16
+
+## Model Overview
+- **Model Architecture:** Qwen/Qwen2.5-VL-3B-Instruct
+  - **Input:** Vision-Text
+  - **Output:** Text
+- **Model Optimizations:**
+  - **Weight quantization:** INT4
+  - **Activation quantization:** FP16
+- **Release Date:** 2/24/2025
+- **Version:** 1.0
+- **Model Developers:** Neural Magic
+
+Quantized version of [Qwen/Qwen2.5-VL-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct).
+
+### Model Optimizations
+
+This model was obtained by quantizing the weights of [Qwen/Qwen2.5-VL-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct) to INT8 data type, ready for inference with vLLM >= 0.5.2.
+
+## Deployment
+
+### Use with vLLM
+
+This model can be deployed efficiently using the [vLLM](https://docs.vllm.ai/en/latest/) backend, as shown in the example below.
+
+```python
+from vllm.assets.image import ImageAsset
+from vllm import LLM, SamplingParams
+
+# prepare model
+llm = LLM(
+    model="neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w4a16",
+    trust_remote_code=True,
+    max_model_len=4096,
+    max_num_seqs=2,
+)
+
+# prepare inputs
+question = "What is the content of this image?"
+inputs = {
+    "prompt": f"<|user|>\n<|image_1|>\n{question}<|end|>\n<|assistant|>\n",
+    "multi_modal_data": {
+        "image": ImageAsset("cherry_blossom").pil_image.convert("RGB")
+    },
+}
+
+# generate response
+print("========== SAMPLE GENERATION ==============")
+outputs = llm.generate(inputs, SamplingParams(temperature=0.2, max_tokens=64))
+print(f"PROMPT  : {outputs[0].prompt}")
+print(f"RESPONSE: {outputs[0].outputs[0].text}")
+print("==========================================")
+```
+
+vLLM also supports OpenAI-compatible serving. See the [documentation](https://docs.vllm.ai/en/latest/) for more details.
+
+## Creation
+
+This model was created with [llm-compressor](https://github.com/vllm-project/llm-compressor) by running the code snippet below as part a multimodal announcement blog.
+
+<details>
+  <summary>Model Creation Code</summary>
+  
+```python
+import base64
+from io import BytesIO
+import torch
+from datasets import load_dataset
+from qwen_vl_utils import process_vision_info
+from transformers import AutoProcessor
+from llmcompressor.modifiers.quantization import GPTQModifier
+from llmcompressor.transformers import oneshot
+from llmcompressor.transformers.tracing import (
+    TraceableQwen2_5_VLForConditionalGeneration,
+)
+from compressed_tensors.quantization import QuantizationArgs, QuantizationType, QuantizationStrategy, ActivationOrdering, QuantizationScheme
+
+# Load model.
+model_id = "Qwen/Qwen2.5-VL-3B-Instruct"
+
+model = TraceableQwen2_5_VLForConditionalGeneration.from_pretrained(
+    model_id,
+    device_map="auto",
+    torch_dtype="auto",
+)
+processor = AutoProcessor.from_pretrained(model_id, trust_remote_code=True)
+
+# Oneshot arguments
+DATASET_ID = "lmms-lab/flickr30k"
+DATASET_SPLIT = {"calibration": "test[:512]"}
+NUM_CALIBRATION_SAMPLES = 512
+MAX_SEQUENCE_LENGTH = 2048
+
+# Load dataset and preprocess.
+ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
+ds = ds.shuffle(seed=42)
+dampening_frac=0.01
+
+# Apply chat template and tokenize inputs.
+def preprocess_and_tokenize(example):
+    # preprocess
+    buffered = BytesIO()
+    example["image"].save(buffered, format="PNG")
+    encoded_image = base64.b64encode(buffered.getvalue())
+    encoded_image_text = encoded_image.decode("utf-8")
+    base64_qwen = f"data:image;base64,{encoded_image_text}"
+    messages = [
+        {
+            "role": "user",
+            "content": [
+                {"type": "image", "image": base64_qwen},
+                {"type": "text", "text": "What does the image show?"},
+            ],
+        }
+    ]
+    text = processor.apply_chat_template(
+        messages, tokenize=False, add_generation_prompt=True
+    )
+    image_inputs, video_inputs = process_vision_info(messages)
+
+    # tokenize
+    return processor(
+        text=[text],
+        images=image_inputs,
+        videos=video_inputs,
+        padding=False,
+        max_length=MAX_SEQUENCE_LENGTH,
+        truncation=True,
+    )
+ds = ds.map(preprocess_and_tokenize, remove_columns=ds["calibration"].column_names)
+
+# Define a oneshot data collator for multimodal inputs.
+def data_collator(batch):
+    assert len(batch) == 1
+    return {key: torch.tensor(value) for key, value in batch[0].items()}
+
+recipe = GPTQModifier(
+    targets="Linear",
+    config_groups={
+        "config_group": QuantizationScheme(
+            targets=["Linear"],
+            weights=QuantizationArgs(
+                num_bits=4,
+                type=QuantizationType.INT,
+                strategy=QuantizationStrategy.GROUP,
+                group_size=128,
+                symmetric=True,
+                dynamic=False,
+                actorder=ActivationOrdering.WEIGHT,
+            ),
+        ),
+    },
+    sequential_targets=["Qwen2_5_VLDecoderLayer"],
+    ignore=["lm_head", "re:visual.*"],
+    update_size=NUM_CALIBRATION_SAMPLES,
+    dampening_frac=dampening_frac
+)
+
+SAVE_DIR=f"{model_id.split('/')[1]}-quantized.w4a16"
+
+# Perform oneshot
+oneshot(
+    model=model,
+    tokenizer=model_id,
+    dataset=ds,
+    recipe=recipe,
+    max_seq_length=MAX_SEQUENCE_LENGTH,
+    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
+    trust_remote_code_model=True,
+    data_collator=data_collator,
+    output_dir=SAVE_DIR
+)
+
+```
+</details>
+
+## Evaluation
+
+The model was evaluated using [mistral-evals](https://github.com/neuralmagic/mistral-evals) for vision-related tasks and using [lm_evaluation_harness](https://github.com/neuralmagic/lm-evaluation-harness) for select text-based benchmarks. The evaluations were conducted using the following commands:
+
+<details>
+<summary>Evaluation Commands</summary>
+  
+### Vision Tasks
+- vqav2
+- docvqa
+- mathvista
+- mmmu
+- chartqa
+
+```
+vllm serve neuralmagic/pixtral-12b-quantized.w8a8 --tensor_parallel_size 1 --max_model_len 25000 --trust_remote_code --max_num_seqs 8 --gpu_memory_utilization 0.9 --dtype float16 --limit_mm_per_prompt image=7
+
+python -m eval.run eval_vllm \
+        --model_name neuralmagic/pixtral-12b-quantized.w8a8 \
+        --url http://0.0.0.0:8000 \
+        --output_dir ~/tmp \
+        --eval_name <vision_task_name>
+```
+
+### Text-based Tasks
+#### MMLU
+  
+```
+lm_eval \
+  --model vllm \
+  --model_args pretrained="<model_name>",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=<n>,gpu_memory_utilization=0.8,enable_chunked_prefill=True,trust_remote_code=True \
+  --tasks mmlu \
+  --num_fewshot 5 \
+  --batch_size auto \
+  --output_path output_dir
+
+```
+
+#### MGSM
+
+```
+lm_eval \
+  --model vllm \
+  --model_args pretrained="<model_name>",dtype=auto,max_model_len=4096,max_gen_toks=2048,max_num_seqs=128,tensor_parallel_size=<n>,gpu_memory_utilization=0.9 \
+  --tasks mgsm_cot_native \
+  --apply_chat_template \
+  --num_fewshot 0 \
+  --batch_size auto \
+  --output_path output_dir
+
+```
+</details>
+
+### Accuracy
+
+<table>
+  <thead>
+    <tr>
+      <th>Category</th>
+      <th>Metric</th>
+      <th>Qwen/Qwen2.5-VL-3B-Instruct</th>
+      <th>Qwen2.5-VL-3B-Instruct-quantized.W4A16</th>
+      <th>Recovery (%)</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td rowspan="6"><b>Vision</b></td>
+      <td>MMMU (val, CoT)<br><i>explicit_prompt_relaxed_correctness</i></td>
+      <td>44.56</td>
+      <td>41.56</td>
+      <td>93.28%</td>
+    </tr>
+    <tr>
+      <td>VQAv2 (val)<br><i>vqa_match</i></td>
+      <td>75.94</td>
+      <td>73.58</td>
+      <td>96.89</td>
+    </tr>
+    <tr>
+      <td>DocVQA (val)<br><i>anls</i></td>
+      <td>92.53</td>
+      <td>91.58</td>
+      <td>98.97%</td>
+    </tr>
+    <tr>
+      <td>ChartQA (test, CoT)<br><i>anywhere_in_answer_relaxed_correctness</i></td>
+      <td>81.20</td>
+      <td>78.96</td>
+      <td>97.24%</td>
+    </tr>
+    <tr>
+      <td>Mathvista (testmini, CoT)<br><i>explicit_prompt_relaxed_correctness</i></td>
+      <td>54.15</td>
+      <td>45.75</td>
+      <td>84.51%</td>
+    </tr>
+    <tr>
+      <td><b>Average Score</b></td>
+      <td><b>69.28</b></td>
+      <td><b>66.29</b></td>
+      <td><b>95.68%</b></td>
+    </tr>
+    <tr>
+      <td rowspan="2"><b>Text</b></td>
+      <td>MGSM (CoT)</td>
+      <td>43.69</td>
+      <td>35.82</td>
+      <td>82.00</td>
+    </tr>
+    <tr>
+      <td>MMLU (5-shot)</td>
+      <td>65.32</td>
+      <td>62.80</td>
+      <td>96.14%</td>
+    </tr>
+  </tbody>
+</table>
+
+
+## Inference Performance
+
+
+This model achieves up to 1.73x speedup in single-stream deployment and up to 3.87x speedup in multi-stream asynchronous deployment, depending on hardware and use-case scenario.
+The following performance benchmarks were conducted with [vLLM](https://docs.vllm.ai/en/latest/) version 0.7.2, and [GuideLLM](https://github.com/neuralmagic/guidellm).
+
+<details>
+<summary>Benchmarking Command</summary>
+```
+  guidellm --model neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w4a16 --target "http://localhost:8000/v1" --data-type emulated --data prompt_tokens=<prompt_tokens>,generated_tokens=<generated_tokens>,images=<num_images>,width=<image_width>,height=<image_height> --max seconds 120 --backend aiohttp_server
+```
+
+</details>
+
+### Single-stream performance (measured with vLLM version 0.7.2)
+
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th style="text-align: center;" colspan="2" >Document Visual Question Answering<br>1680W x 2240H<br>64/128</th>
+      <th style="text-align: center;" colspan="2" >Visual Reasoning <br>640W x 480H<br>128/128</th>
+      <th style="text-align: center;" colspan="2" >Image Captioning<br>480W x 360H<br>0/128</th>
+    </tr>
+    <tr>
+      <th>Hardware</th>
+      <th>Model</th>
+      <th>Average Cost Reduction</th>
+      <th>Latency (s)</th>
+      <th>Queries Per Dollar</th>
+      <th>Latency (s)th>
+      <th>Queries Per Dollar</th>
+      <th>Latency (s)</th>
+      <th>Queries Per Dollar</th>
+    </tr>
+  </thead>
+  <tbody style="text-align: center">
+    <tr>
+      <th rowspan="3" valign="top">A6000x1</th>
+      <th>Qwen/Qwen2.5-VL-3B-Instruct</th>
+      <td></td>
+      <td>3.1</td>
+      <td>1454</td>
+      <td>1.8</td>
+      <td>2546</td>
+      <td>1.7</td>
+      <td>2610</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w8a8</th>
+      <td>1.27</td>
+      <td>2.6</td>
+      <td>1708</td>
+      <td>1.3</td>
+      <td>3340</td>
+      <td>1.3</td>
+      <td>3459</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w4a16</th>
+      <td>1.57</td>
+      <td>2.4</td>
+      <td>1886</td>
+      <td>1.0</td>
+      <td>4409</td>
+      <td>1.0</td>
+      <td>4409</td>
+    </tr>
+    <tr>
+      <th rowspan="3" valign="top">A100x1</th>
+      <th>Qwen/Qwen2.5-VL-3B-Instruct</th>
+      <td></td>
+      <td>2.2</td>
+      <td>920</td>
+      <td>1.3</td>
+      <td>1603</td>
+      <td>1.2</td>
+      <td>1636</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w8a8</th>
+      <td>1.09</td>
+      <td>2.1</td>
+      <td>975</td>
+      <td>1.2</td>
+      <td>1743</td>
+      <td>1.1</td>
+      <td>1814</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w4a16</th>
+      <td>1.20</td>
+      <td>2.0</td>
+      <td>1011</td>
+      <td>1.0</td>
+      <td>2015</td>
+      <td>1.0</td>
+      <td>2012</td>
+    </tr>
+    <tr>
+      <th rowspan="3" valign="top">H100x1</th>
+      <th>Qwen/Qwen2.5-VL-3B-Instruct</th>
+      <td></td>
+      <td>1.5</td>
+      <td>740</td>
+      <td>0.9</td>
+      <td>1221</td>
+      <td>0.9</td>
+      <td>1276</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-FP8-Dynamic</th>
+      <td>1.06</td>
+      <td>1.4</td>
+      <td>768</td>
+      <td>0.9</td>
+      <td>1276</td>
+      <td>0.8</td>
+      <td>1399</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w4a16</th>
+      <td>1.24</td>
+      <td>0.9</td>
+      <td>1219</td>
+      <td>0.9</td>
+      <td>1270</td>
+      <td>0.8</td>
+      <td>1304</td>
+    </tr>
+  </tbody>
+</table>
+ 
+**Use case profiles: Image Size (WxH) / prompt tokens / generation tokens
+
+**QPD: Queries per dollar, based on on-demand cost at [Lambda Labs](https://lambdalabs.com/service/gpu-cloud) (observed on 2/18/2025).
+
+### Multi-stream asynchronous performance (measured with vLLM version 0.7.2)
+
+<table border="1" class="dataframe">
+  <thead>
+    <tr>
+      <th></th>
+      <th></th>
+      <th></th>
+      <th style="text-align: center;" colspan="2" >Document Visual Question Answering<br>1680W x 2240H<br>64/128</th>
+      <th style="text-align: center;" colspan="2" >Visual Reasoning <br>640W x 480H<br>128/128</th>
+      <th style="text-align: center;" colspan="2" >Image Captioning<br>480W x 360H<br>0/128</th>
+    </tr>
+    <tr>
+      <th>Hardware</th>
+      <th>Model</th>
+      <th>Average Cost Reduction</th>
+      <th>Maximum throughput (QPS)</th>
+      <th>Queries Per Dollar</th>
+      <th>Maximum throughput (QPS)</th>
+      <th>Queries Per Dollar</th>
+      <th>Maximum throughput (QPS)</th>
+      <th>Queries Per Dollar</th>
+    </tr>
+  </thead>
+  <tbody style="text-align: center">
+    <tr>
+      <th rowspan="3" valign="top">A6000x1</th>
+      <th>Qwen/Qwen2.5-VL-3B-Instruct</th>
+      <td></td>
+      <td>0.5</td>
+      <td>2405</td>
+      <td>2.6</td>
+      <td>11889</td>
+      <td>2.9</td>
+      <td>12909</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w8a8</th>
+      <td>1.26</td>
+      <td>0.6</td>
+      <td>2725</td>
+      <td>3.4</td>
+      <td>15162</td>
+      <td>3.9</td>
+      <td>17673</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w4a16</th>
+      <td>1.39</td>
+      <td>0.6</td>
+      <td>2548</td>
+      <td>3.9</td>
+      <td>17437</td>
+      <td>4.7</td>
+      <td>21223</td>
+    </tr>
+    <tr>
+      <th rowspan="3" valign="top">A100x1</th>
+      <th>Qwen/Qwen2.5-VL-3B-Instruct</th>
+      <td></td>
+      <td>0.8</td>
+      <td>1663</td>
+      <td>3.9</td>
+      <td>7899</td>
+      <td>4.4</td>
+      <td>8924</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w8a8</th>
+      <td>1.06</td>
+      <td>0.9</td>
+      <td>1734</td>
+      <td>4.2</td>
+      <td>8488</td>
+      <td>4.7</td>
+      <td>9548</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w4a16</th>
+      <td>1.10</td>
+      <td>0.9</td>
+      <td>1775</td>
+      <td>4.2</td>
+      <td>8540</td>
+      <td>5.1</td>
+      <td>10318</td>
+    </tr>
+    <tr>
+      <th rowspan="3" valign="top">H100x1</th>
+      <th>Qwen/Qwen2.5-VL-3B-Instruct</th>
+      <td></td>
+      <td>1.1</td>
+      <td>1188</td>
+      <td>4.3</td>
+      <td>4656</td>
+      <td>4.3</td>
+      <td>4676</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-FP8-Dynamic</th>
+      <td>1.15</td>
+      <td>1.4</td>
+      <td>1570</td>
+      <td>4.3</td>
+      <td>4676</td>
+      <td>4.8</td>
+      <td>5220</td>
+    </tr>
+    <tr>
+      <th>neuralmagic/Qwen2.5-VL-3B-Instruct-quantized.w4a16</th>
+      <td>1.96</td>
+      <td>4.2</td>
+      <td>4598</td>
+      <td>4.1</td>
+      <td>4505</td>
+      <td>4.4</td>
+      <td>4838</td>
+    </tr>
+  </tbody>
+</table>
+
+**Use case profiles: Image Size (WxH) / prompt tokens / generation tokens
+
+**QPS: Queries per second.
+
+**QPD: Queries per dollar, based on on-demand cost at [Lambda Labs](https://lambdalabs.com/service/gpu-cloud) (observed on 2/18/2025).
--- a/added_tokens.json
+++ b/added_tokens.json
@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}
--- a/chat_template.json
+++ b/chat_template.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94174d7176c52a7192f96fc34eb2cf23c7c2059d63cdbfadca1586ba89731fb7
+size 1049
--- a/config.json
+++ b/config.json
@@ -0,0 +1,244 @@
+{
+  "_name_or_path": "Qwen/Qwen2.5-VL-3B-Instruct",
+  "architectures": [
+    "Qwen2_5_VLForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "image_token_id": 151655,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 128000,
+  "max_window_layers": 70,
+  "model_type": "qwen2_5_vl",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "quantization_config": {
+    "config_groups": {
+      "group_0": {
+        "input_activations": null,
+        "output_activations": null,
+        "targets": [
+          "Linear"
+        ],
+        "weights": {
+          "actorder": "weight",
+          "block_structure": null,
+          "dynamic": false,
+          "group_size": 128,
+          "num_bits": 4,
+          "observer": "minmax",
+          "observer_kwargs": {},
+          "strategy": "group",
+          "symmetric": true,
+          "type": "int"
+        }
+      }
+    },
+    "format": "pack-quantized",
+    "global_compression_ratio": 1.3990843687790238,
+    "ignore": [
+      "visual.blocks.0.attn.qkv",
+      "visual.blocks.0.attn.proj",
+      "visual.blocks.0.mlp.gate_proj",
+      "visual.blocks.0.mlp.up_proj",
+      "visual.blocks.0.mlp.down_proj",
+      "visual.blocks.1.attn.qkv",
+      "visual.blocks.1.attn.proj",
+      "visual.blocks.1.mlp.gate_proj",
+      "visual.blocks.1.mlp.up_proj",
+      "visual.blocks.1.mlp.down_proj",
+      "visual.blocks.2.attn.qkv",
+      "visual.blocks.2.attn.proj",
+      "visual.blocks.2.mlp.gate_proj",
+      "visual.blocks.2.mlp.up_proj",
+      "visual.blocks.2.mlp.down_proj",
+      "visual.blocks.3.attn.qkv",
+      "visual.blocks.3.attn.proj",
+      "visual.blocks.3.mlp.gate_proj",
+      "visual.blocks.3.mlp.up_proj",
+      "visual.blocks.3.mlp.down_proj",
+      "visual.blocks.4.attn.qkv",
+      "visual.blocks.4.attn.proj",
+      "visual.blocks.4.mlp.gate_proj",
+      "visual.blocks.4.mlp.up_proj",
+      "visual.blocks.4.mlp.down_proj",
+      "visual.blocks.5.attn.qkv",
+      "visual.blocks.5.attn.proj",
+      "visual.blocks.5.mlp.gate_proj",
+      "visual.blocks.5.mlp.up_proj",
+      "visual.blocks.5.mlp.down_proj",
+      "visual.blocks.6.attn.qkv",
+      "visual.blocks.6.attn.proj",
+      "visual.blocks.6.mlp.gate_proj",
+      "visual.blocks.6.mlp.up_proj",
+      "visual.blocks.6.mlp.down_proj",
+      "visual.blocks.7.attn.qkv",
+      "visual.blocks.7.attn.proj",
+      "visual.blocks.7.mlp.gate_proj",
+      "visual.blocks.7.mlp.up_proj",
+      "visual.blocks.7.mlp.down_proj",
+      "visual.blocks.8.attn.qkv",
+      "visual.blocks.8.attn.proj",
+      "visual.blocks.8.mlp.gate_proj",
+      "visual.blocks.8.mlp.up_proj",
+      "visual.blocks.8.mlp.down_proj",
+      "visual.blocks.9.attn.qkv",
+      "visual.blocks.9.attn.proj",
+      "visual.blocks.9.mlp.gate_proj",
+      "visual.blocks.9.mlp.up_proj",
+      "visual.blocks.9.mlp.down_proj",
+      "visual.blocks.10.attn.qkv",
+      "visual.blocks.10.attn.proj",
+      "visual.blocks.10.mlp.gate_proj",
+      "visual.blocks.10.mlp.up_proj",
+      "visual.blocks.10.mlp.down_proj",
+      "visual.blocks.11.attn.qkv",
+      "visual.blocks.11.attn.proj",
+      "visual.blocks.11.mlp.gate_proj",
+      "visual.blocks.11.mlp.up_proj",
+      "visual.blocks.11.mlp.down_proj",
+      "visual.blocks.12.attn.qkv",
+      "visual.blocks.12.attn.proj",
+      "visual.blocks.12.mlp.gate_proj",
+      "visual.blocks.12.mlp.up_proj",
+      "visual.blocks.12.mlp.down_proj",
+      "visual.blocks.13.attn.qkv",
+      "visual.blocks.13.attn.proj",
+      "visual.blocks.13.mlp.gate_proj",
+      "visual.blocks.13.mlp.up_proj",
+      "visual.blocks.13.mlp.down_proj",
+      "visual.blocks.14.attn.qkv",
+      "visual.blocks.14.attn.proj",
+      "visual.blocks.14.mlp.gate_proj",
+      "visual.blocks.14.mlp.up_proj",
+      "visual.blocks.14.mlp.down_proj",
+      "visual.blocks.15.attn.qkv",
+      "visual.blocks.15.attn.proj",
+      "visual.blocks.15.mlp.gate_proj",
+      "visual.blocks.15.mlp.up_proj",
+      "visual.blocks.15.mlp.down_proj",
+      "visual.blocks.16.attn.qkv",
+      "visual.blocks.16.attn.proj",
+      "visual.blocks.16.mlp.gate_proj",
+      "visual.blocks.16.mlp.up_proj",
+      "visual.blocks.16.mlp.down_proj",
+      "visual.blocks.17.attn.qkv",
+      "visual.blocks.17.attn.proj",
+      "visual.blocks.17.mlp.gate_proj",
+      "visual.blocks.17.mlp.up_proj",
+      "visual.blocks.17.mlp.down_proj",
+      "visual.blocks.18.attn.qkv",
+      "visual.blocks.18.attn.proj",
+      "visual.blocks.18.mlp.gate_proj",
+      "visual.blocks.18.mlp.up_proj",
+      "visual.blocks.18.mlp.down_proj",
+      "visual.blocks.19.attn.qkv",
+      "visual.blocks.19.attn.proj",
+      "visual.blocks.19.mlp.gate_proj",
+      "visual.blocks.19.mlp.up_proj",
+      "visual.blocks.19.mlp.down_proj",
+      "visual.blocks.20.attn.qkv",
+      "visual.blocks.20.attn.proj",
+      "visual.blocks.20.mlp.gate_proj",
+      "visual.blocks.20.mlp.up_proj",
+      "visual.blocks.20.mlp.down_proj",
+      "visual.blocks.21.attn.qkv",
+      "visual.blocks.21.attn.proj",
+      "visual.blocks.21.mlp.gate_proj",
+      "visual.blocks.21.mlp.up_proj",
+      "visual.blocks.21.mlp.down_proj",
+      "visual.blocks.22.attn.qkv",
+      "visual.blocks.22.attn.proj",
+      "visual.blocks.22.mlp.gate_proj",
+      "visual.blocks.22.mlp.up_proj",
+      "visual.blocks.22.mlp.down_proj",
+      "visual.blocks.23.attn.qkv",
+      "visual.blocks.23.attn.proj",
+      "visual.blocks.23.mlp.gate_proj",
+      "visual.blocks.23.mlp.up_proj",
+      "visual.blocks.23.mlp.down_proj",
+      "visual.blocks.24.attn.qkv",
+      "visual.blocks.24.attn.proj",
+      "visual.blocks.24.mlp.gate_proj",
+      "visual.blocks.24.mlp.up_proj",
+      "visual.blocks.24.mlp.down_proj",
+      "visual.blocks.25.attn.qkv",
+      "visual.blocks.25.attn.proj",
+      "visual.blocks.25.mlp.gate_proj",
+      "visual.blocks.25.mlp.up_proj",
+      "visual.blocks.25.mlp.down_proj",
+      "visual.blocks.26.attn.qkv",
+      "visual.blocks.26.attn.proj",
+      "visual.blocks.26.mlp.gate_proj",
+      "visual.blocks.26.mlp.up_proj",
+      "visual.blocks.26.mlp.down_proj",
+      "visual.blocks.27.attn.qkv",
+      "visual.blocks.27.attn.proj",
+      "visual.blocks.27.mlp.gate_proj",
+      "visual.blocks.27.mlp.up_proj",
+      "visual.blocks.27.mlp.down_proj",
+      "visual.blocks.28.attn.qkv",
+      "visual.blocks.28.attn.proj",
+      "visual.blocks.28.mlp.gate_proj",
+      "visual.blocks.28.mlp.up_proj",
+      "visual.blocks.28.mlp.down_proj",
+      "visual.blocks.29.attn.qkv",
+      "visual.blocks.29.attn.proj",
+      "visual.blocks.29.mlp.gate_proj",
+      "visual.blocks.29.mlp.up_proj",
+      "visual.blocks.29.mlp.down_proj",
+      "visual.blocks.30.attn.qkv",
+      "visual.blocks.30.attn.proj",
+      "visual.blocks.30.mlp.gate_proj",
+      "visual.blocks.30.mlp.up_proj",
+      "visual.blocks.30.mlp.down_proj",
+      "visual.blocks.31.attn.qkv",
+      "visual.blocks.31.attn.proj",
+      "visual.blocks.31.mlp.gate_proj",
+      "visual.blocks.31.mlp.up_proj",
+      "visual.blocks.31.mlp.down_proj",
+      "visual.merger.mlp.0",
+      "visual.merger.mlp.2",
+      "lm_head"
+    ],
+    "kv_cache_scheme": null,
+    "quant_method": "compressed-tensors",
+    "quantization_status": "compressed",
+    "sparsity_config": {}
+  },
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": {
+    "mrope_section": [
+      16,
+      24,
+      24
+    ],
+    "rope_type": "default",
+    "type": "default"
+  },
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.49.0.dev0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "video_token_id": 151656,
+  "vision_config": {
+    "hidden_size": 1280,
+    "in_chans": 3,
+    "model_type": "qwen2_5_vl",
+    "out_hidden_size": 2048,
+    "spatial_patch_size": 14,
+    "tokens_per_second": 2
+  },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652,
+  "vision_token_id": 151654,
+  "vocab_size": 151936
+}
--- a/configuration.json
+++ b/configuration.json
@@ -0,0 +1 @@
+{"framework": "pytorch", "task": "image-to-text", "allow_remote": true}
--- a/generation_config.json
+++ b/generation_config.json
@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 0.1,
+  "top_k": 1,
+  "top_p": 0.001,
+  "transformers_version": "4.49.0.dev0"
+}
--- a/merges.txt
+++ b/merges.txt
--- a/model.safetensors
+++ b/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1811c5fa738747144c60821401a200d4e56339d80377c0c497eb3b341d85ff23
+size 4013282536
--- a/preprocessor_config.json
+++ b/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": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "Qwen2_5_VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 12845056,
+    "shortest_edge": 3136
+  },
+  "temporal_patch_size": 2
+}
--- a/recipe.yaml
+++ b/recipe.yaml
@@ -0,0 +1,12 @@
+DEFAULT_stage:
+  DEFAULT_modifiers:
+    GPTQModifier:
+      sequential_targets: [Qwen2_5_VLDecoderLayer]
+      dampening_frac: 0.01
+      config_groups:
+        config_group:
+          targets: [Linear]
+          weights: {num_bits: 4, type: int, symmetric: true, group_size: 128, strategy: group,
+            dynamic: false, actorder: weight, observer: minmax}
+      targets: Linear
+      ignore: [lm_head, 're:visual.*']
--- a/special_tokens_map.json
+++ b/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
+  }
+}
--- a/tokenizer.json
+++ b/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896
--- a/tokenizer_config.json
+++ b/tokenizer_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c182cc5f590489eba7c5435864b0814fcd9256ef72933f8ccc142116974dbb01
+size 7300
--- a/vocab.json
+++ b/vocab.json
				`@@ -0,0 +1 @@`
				`{"framework": "pytorch", "task": "image-to-text", "allow_remote": true}`