Upload 12 files

.gitattributes

@@ -46,3 +46,8 @@ assets/11111.png filter=lfs diff=lfs merge=lfs -text
 assets/22222.png filter=lfs diff=lfs merge=lfs -text
 assets/33333.png filter=lfs diff=lfs merge=lfs -text
 assets/44444.png filter=lfs diff=lfs merge=lfs -text
+example_images/campeones.jpg filter=lfs diff=lfs merge=lfs -text
+example_images/document.jpg filter=lfs diff=lfs merge=lfs -text
+example_images/examples_weather_events.png filter=lfs diff=lfs merge=lfs -text
+example_images/int.png filter=lfs diff=lfs merge=lfs -text
+example_images/newyork.jpg filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,14 @@
+---
+title: Qwen2.5 VL 3B
+emoji: 🔥
+colorFrom: pink
+colorTo: indigo
+sdk: gradio
+sdk_version: 5.13.1
+app_file: app.py
+pinned: true
+license: creativeml-openrail-m
+short_description: Qwen2.5-VL-3B-Instruct
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,98 @@
+import gradio as gr
+from transformers import AutoProcessor, Qwen2_5_VLForConditionalGeneration, TextIteratorStreamer
+from transformers.image_utils import load_image
+from threading import Thread
+import time
+import torch
+import spaces
+
+MODEL_ID = "Qwen/Qwen2.5-VL-3B-Instruct"
+processor = AutoProcessor.from_pretrained(MODEL_ID, trust_remote_code=True)
+model = Qwen2_5_VLForConditionalGeneration.from_pretrained(
+    MODEL_ID,
+    trust_remote_code=True,
+    torch_dtype=torch.bfloat16
+).to("cuda").eval()
+
+@spaces.GPU
+def model_inference(input_dict, history):
+    text = input_dict["text"]
+    files = input_dict["files"]
+
+    # Load images if provided
+    if len(files) > 1:
+        images = [load_image(image) for image in files]
+    elif len(files) == 1:
+        images = [load_image(files[0])]
+    else:
+        images = []
+
+    # Validate input
+    if text == "" and not images:
+        gr.Error("Please input a query and optionally image(s).")
+        return
+    if text == "" and images:
+        gr.Error("Please input a text query along with the image(s).")
+        return
+
+    # Prepare messages for the model
+    messages = [
+        {
+            "role": "user",
+            "content": [
+                *[{"type": "image", "image": image} for image in images],
+                {"type": "text", "text": text},
+            ],
+        }
+    ]
+
+    # Apply chat template and process inputs
+    prompt = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+    inputs = processor(
+        text=[prompt],
+        images=images if images else None,
+        return_tensors="pt",
+        padding=True,
+    ).to("cuda")
+
+    # Set up streamer for real-time output
+    streamer = TextIteratorStreamer(processor, skip_prompt=True, skip_special_tokens=True)
+    generation_kwargs = dict(inputs, streamer=streamer, max_new_tokens=1024)
+
+    # Start generation in a separate thread
+    thread = Thread(target=model.generate, kwargs=generation_kwargs)
+    thread.start()
+
+    # Stream the output
+    buffer = ""
+    yield "Thinking..."
+    for new_text in streamer:
+        buffer += new_text
+        time.sleep(0.01)
+        yield buffer
+
+
+# Example inputs
+examples = [
+    [{"text": "Explain the movie scene; screen board", "files": ["example_images/int.png"]}],
+    [{"text": "Describe the document?", "files": ["example_images/document.jpg"]}],
+    [{"text": "Describe this image.", "files": ["example_images/campeones.jpg"]}],
+    [{"text": "What does this say?", "files": ["example_images/math.jpg"]}],
+    [{"text": "What is this UI about?", "files": ["example_images/s2w_example.png"]}],
+    [{"text": "Can you describe this image?", "files": ["example_images/newyork.jpg"]}],
+    [{"text": "Can you describe this image?", "files": ["example_images/dogs.jpg"]}],
+    [{"text": "Where do the severe droughts happen according to this diagram?", "files": ["example_images/examples_weather_events.png"]}],
+
+]
+
+demo = gr.ChatInterface(
+    fn=model_inference,
+    description="# **Qwen2.5-VL-3B-Instruct**",
+    examples=examples,
+    textbox=gr.MultimodalTextbox(label="Query Input", file_types=["image"], file_count="multiple"),
+    stop_btn="Stop Generation",
+    multimodal=True,
+    cache_examples=False,
+)
+
+demo.launch(debug=True)

requirements.txt

@@ -0,0 +1,8 @@
+gradio_client==1.3.0
+qwen-vl-utils==0.0.2
+transformers-stream-generator==0.0.4
+torch==2.4.0
+torchvision==0.19.0
+git+https://github.com/huggingface/transformers.git
+accelerate
+av