Oculus 0.1
Hybrid-reasoning vision-language model built on the Oceanir-Oculus OO1 Architecture.
Small models that outperform systems 10x larger on visual reasoning and perception tasks, running on commodity GPUs or edge devices.
What's New in Oculus 0.1
Reasoning via Thinking Traces
Short, structured reasoning traces improve multi-step decisions, small-object understanding, and ambiguous spatial tasks.
1 answer = model . ask ( image , "How many red cars on the left?" , think = True )
2 # Output includes <think>...</think> reasoning trace
Perceptive Tool Calling + Focus (Zoom & Crop)
Oculus can trigger tool calls to focus (zoom and crop) and re-query on smaller regions — dramatically improving fine-grained perception.
1 answer = model . ask ( image , "Read the small text on the sign" , focus = True )
2 # Model automatically zooms to relevant region
Structured Outputs
More reliable structured output generation for consistent JSON and predictable downstream integration.
1 result = model . generate ( image , prompt = "List all objects" , mode = "json" )
2 # Returns structured JSON: {"objects": [{"label": "car", "box": [x1,y1,x2,y2]}, ...]}
Complex OCR
Improved text recognition across cluttered, low-resolution, or distorted regions — enabling accurate extraction from documents, diagrams, labels, screens, and dense real-world scenes.
text = model.ocr(image) # Extracts text from any visual contentDesktop Use
Better performance on everyday desktop and mobile workflows such as UI understanding and navigation, making Oculus faster and more capable for agentic use cases.
1 elements = model . detect_ui ( screenshot )
2 # Returns: [{"type": "button", "text": "Submit", "bbox": [x1,y1,x2,y2]}, ...]
Architecture
Oceanir-Oculus OO1 Architecture — A hybrid vision-language architecture optimized for:
Visual reasoning outperforming systems 10x larger
Edge deployment on commodity GPUs
Grounded perception with spatial understanding
Tool calling and agentic workflows
Installation
Usage
1 from oceanir import Oculus
2
3 model = Oculus . from_pretrained ( "OceanirAI/Oculus-0.1" )
4
5 # Basic VQA
6 answer = model . ask ( "image.jpg" , "What is this?" )
7
8 # With reasoning traces
9 answer = model . ask ( "scene.jpg" , "Count the people" , think = True )
10
11 # With focus/zoom for fine details
12 answer = model . ask ( "document.jpg" , "Read the fine print" , focus = True )
13
14 # Structured JSON output
15 result = model . generate ( image , prompt = "Describe objects" , mode = "json" )
16
17 # OCR
18 text = model . ocr ( "screenshot.png" )
19
20 # UI Detection
21 ui_elements = model . detect_ui ( "desktop.png" )
22
23 # Object Detection with grounding
24 boxes = model . detect ( "image.jpg" )
25
26 # Segmentation
27 mask = model . segment ( "image.jpg" )
Output Modes
Mode Method Output Text model.ask(image, question)Natural language answer Reasoning model.ask(image, question, think=True)Answer with <think> trace JSON model.generate(image, mode="json")Structured JSON Points model.generate(image, mode="point")Object center points Boxes model.detect(image)Bounding boxes + labels Polygons model.segment(image)Segmentation masks OCR model.ocr(image)Extracted text + locations UI model.detect_ui(image)UI elements + types
Special Tokens
Token Purpose <think>...</think>Reasoning traces <focus>...</focus>Focus/zoom regions <json>...</json>Structured output <box>...</box>Bounding box coordinates <point>...</point>Point coordinates
Use Cases
Robotics : Grounded perception for manipulation and navigationIndustrial Inspection : Defect detection and quality controlDocument Processing : Complex OCR and form extractionMedia Search : Visual content understanding and retrievalDesktop Automation : UI understanding for agentic workflowsSecurity : Visual monitoring and anomaly detection
What's in This Repo
trained_components/projector.npz - Vision-language projectortrained_components/heads.pth - Task heads (detection, segmentation, OCR, UI)oculus_unified_model/ - Model code
License
Oceanir Research License - Non-commercial research only.
For commercial licensing:
licensing@oceanir.ai 
