Instructions to use prithivMLmods/WASP-2B-VL-Highlights with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use prithivMLmods/WASP-2B-VL-Highlights with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="prithivMLmods/WASP-2B-VL-Highlights")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModel

processor = AutoProcessor.from_pretrained("prithivMLmods/WASP-2B-VL-Highlights")
model = AutoModel.from_pretrained("prithivMLmods/WASP-2B-VL-Highlights")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use prithivMLmods/WASP-2B-VL-Highlights with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "prithivMLmods/WASP-2B-VL-Highlights"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prithivMLmods/WASP-2B-VL-Highlights",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/prithivMLmods/WASP-2B-VL-Highlights

SGLang

How to use prithivMLmods/WASP-2B-VL-Highlights with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "prithivMLmods/WASP-2B-VL-Highlights" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prithivMLmods/WASP-2B-VL-Highlights",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "prithivMLmods/WASP-2B-VL-Highlights" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prithivMLmods/WASP-2B-VL-Highlights",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use prithivMLmods/WASP-2B-VL-Highlights with Docker Model Runner:
```
docker model run hf.co/prithivMLmods/WASP-2B-VL-Highlights
```

Update README.md

by prithivMLmods - opened May 2, 2025

base: refs/heads/main

←

from: refs/pr/2

Discussion Files changed

+17

-17

Files changed (1) hide show

README.md +17 -17

README.md CHANGED Viewed

@@ -7,39 +7,39 @@ base_model:
 pipeline_tag: image-text-to-text
 library_name: transformers
 tags:
-- Highlights
 - Generation
 - OCR
 - KIE
 ---
-![fxghdfgh.png](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/svbsK16pmYR9Q4FoFHNYs.png)
-# **Needle-2B-VL-Highlights**
->  [!Note]
->  The **Needle-2B-VL-Highlights** model is a fine-tuned version of *Qwen2-VL-2B-Instruct*, specifically optimized for **image highlights extraction**, **messy handwriting recognition**, **Optical Character Recognition (OCR)**, **English language understanding**, and **math problem solving with LaTeX formatting**. This model uses a conversational visual-language interface to effectively handle multi-modal tasks.
-[![Open Demo in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/#fileId=https%3A//huggingface.co/prithivMLmods/Needle-2B-VL-Highlights/blob/main/Callisto_OCR3_2B_Instruct.ipynb)
 # **Key Enhancements:**
-* **State-of-the-art image comprehension** across varying resolutions and aspect ratios:
-  Needle-2B-VL-Highlights delivers top-tier performance on benchmarks such as MathVista, DocVQA, RealWorldQA, and MTVQA.
-* **Image Highlighting Expertise**:
   Specially tuned to **identify and summarize key visual elements** in an image — ideal for **creating visual highlights**, annotations, and summaries.
-* **Handwriting OCR Enhanced**:
   Recognizes **messy and complex handwritten notes** with precision, perfect for digitizing real-world documents.
-* **Video Content Understanding**:
-  Capable of processing videos longer than 20 minutes for **context-aware Q&A, transcription**, and **highlight extraction**.
-* **Multi-device Integration**:
   Can be used as an intelligent agent for mobile phones, robots, and other devices — able to **understand visual scenes and execute actions**.
-* **Multilingual OCR Support**:
   In addition to English and Chinese, supports OCR for European languages, Japanese, Korean, Arabic, and Vietnamese.
 # **Run with Transformers🤗**
@@ -76,7 +76,7 @@ from docx.enum.text import WD_ALIGN_PARAGRAPH
 # Define model options
 MODEL_OPTIONS = {
-    "Needle-2B-VL-Highlights": "prithivMLmods/Needle-2B-VL-Highlights",
 }
 # Preload models and processors into CUDA
@@ -288,7 +288,7 @@ with gr.Blocks(css=css) as demo:
                 model_choice = gr.Dropdown(
                     label="Model Selection",
                     choices=list(MODEL_OPTIONS.keys()),
-                    value="Needle-2B-VL-Highlights"
                 )
                 input_media = gr.File(
                     label="Upload Image", type="filepath"

 pipeline_tag: image-text-to-text
 library_name: transformers
 tags:
 - Generation
 - OCR
 - KIE
+- Highlights-Generator
 ---
+![WASP.png](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/BpmMrx7Vsm3Pnfqb2xGxC.png)
+ # **WASP-2B-VL-Highlights**
+> \[!Note]
+> The **WASP-2B-VL-Highlights** model is a fine-tuned version of *Qwen2-VL-2B-Instruct*, specifically optimized for **image highlights extraction**, **messy handwriting recognition**, **Optical Character Recognition (OCR)**, **English language understanding**, and **math problem solving with LaTeX formatting**. This model uses a conversational visual-language interface to effectively handle multi-modal tasks.
+[![Open Demo in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/#fileId=https%3A//huggingface.co/prithivMLmods/WASP-2B-VL-Highlights/blob/main/Callisto_OCR3_2B_Instruct.ipynb)
 # **Key Enhancements:**
+* **State-of-the-art image comprehension** across varying resolutions and aspect ratios:
+  WASP-2B-VL-Highlights delivers top-tier performance on benchmarks such as MathVista, DocVQA, RealWorldQA, and MTVQA.
+* **Image Highlighting Expertise**:
   Specially tuned to **identify and summarize key visual elements** in an image — ideal for **creating visual highlights**, annotations, and summaries.
+* **Handwriting OCR Enhanced**:
   Recognizes **messy and complex handwritten notes** with precision, perfect for digitizing real-world documents.
+* **Video Content Understanding**:
+  Capable of processing videos longer than 20 minutes for **context-aware Q\&A, transcription**, and **highlight extraction**.
+* **Multi-device Integration**:
   Can be used as an intelligent agent for mobile phones, robots, and other devices — able to **understand visual scenes and execute actions**.
+* **Multilingual OCR Support**:
   In addition to English and Chinese, supports OCR for European languages, Japanese, Korean, Arabic, and Vietnamese.
 # **Run with Transformers🤗**
 # Define model options
 MODEL_OPTIONS = {
+    "Needle-2B-VL-Highlights": "prithivMLmods/WASP-2B-VL-Highlights",
 }
 # Preload models and processors into CUDA
                 model_choice = gr.Dropdown(
                     label="Model Selection",
                     choices=list(MODEL_OPTIONS.keys()),
+                    value="WASP-2B-VL-Highlights"
                 )
                 input_media = gr.File(
                     label="Upload Image", type="filepath"