Sold by: Woven by Toyota, Inc.
Deployed on AWS
A multi-modal Large Language Model (LLM) that processes text and images/videos simultaneously. Our model is deployed on Amazon SageMaker for inference, with support for visual question answering, spatial-temporal content analysis, image/video analysis and captioning, etc.
Overview
Woven City AI Vision Engine is a multi-modal LLM that focuses on the spatial-temporal understanding from the image/video data. This model enables users to analyze and understand content across text and images/videos in a single, unified interface.
To best support your user experience, please fill out this form before subscribing our model:
https://forms.gle/TZSdUJxLuurLHb2K6
Key Capabilities
- Support image and video understanding
- Support short and long videos
- Top level performance on a public spatial-temporal understanding benchmark
Industry Applications
- Retail & marketing visual analysis
- Facility management / security video understanding / activity anticipation
- Manufacturing visual inspection / activity understanding
- Mobility Vision for safety
Enterprise Features
- AWS IAM integration for secure access control
- Comprehensive audit logging and monitoring
- Scalable deployment options from development to production
Deployment
Deployment ready on Amazon SageMaker with comprehensive API documentation, sample notebooks, and best practices guidance for immediate integration into existing workflows.
Highlights
- Hierarchical Compression: Employs a sophisticated two-tier compression strategy. A fine-grained compressor that processes short-term temporal information, while a coarse-grained compressor handles long-term context.
- Long-Context Understanding: Capable of processing and understanding videos of extended duration.
- Top-level performance: Outperforms existing models in long-video question answering and excels in the video understanding benchmark MVBench with a top level score.
Details
Sold by
Categories
Delivery method
Latest version
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.g5.12xlarge Inference (Batch) Recommended | Model inference on the ml.g5.12xlarge instance type, batch mode | $6.00 |
ml.g5.12xlarge Inference (Real-Time) Recommended | Model inference on the ml.g5.12xlarge instance type, real-time mode | $6.00 |
ml.g5.8xlarge Inference (Batch) | Model inference on the ml.g5.8xlarge instance type, batch mode | $5.00 |
ml.g5.24xlarge Inference (Batch) | Model inference on the ml.g5.24xlarge instance type, batch mode | $7.00 |
ml.g5.48xlarge Inference (Batch) | Model inference on the ml.g5.48xlarge instance type, batch mode | $8.00 |
ml.g5.8xlarge Inference (Real-Time) | Model inference on the ml.g5.8xlarge instance type, real-time mode | $5.00 |
ml.g5.24xlarge Inference (Real-Time) | Model inference on the ml.g5.24xlarge instance type, real-time mode | $7.00 |
ml.g5.48xlarge Inference (Real-Time) | Model inference on the ml.g5.48xlarge instance type, real-time mode | $8.00 |
Vendor refund policy
No refund
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
Amazon SageMaker model
An Amazon SageMaker model package is a pre-trained machine learning model ready to use without additional training. Use the model package to create a model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Deploy the model on Amazon SageMaker AI using the following options:
Version release notes
Woven City AI Vision Engine v1.0.5-alpha - Initial Release
This release introduces a powerful multimodal large language model capable of understanding and analyzing text, images, and videos. This model delivers enterprise-grade performance for visual AI applications.
New Features:
- Multimodal Understanding: Process text, images, and videos in a single request
- Easy-to-use Inference: Deploy on Amazon SageMaker endpoints for low-latency responses
- Multi-turn Conversations: Maintain context across multiple interactions with chat history support
- High-Quality Image Analysis: Support for JPG, PNG
- High-Quality Video Analysis: Support for MP4 (recommended), AVI, with H.264/MPEG-4 codecs preferred
- Configurable Generation: Fine-tune responses with temperature, top-p, top-k, and beam search parameters
- Production-Ready: Built-in error handling, rate limiting, and comprehensive logging
Additional details
Inputs
- Summary
The model accepts JSON requests with multimodal content including text prompts, images, and videos for comprehensive analysis and understanding.
- Input MIME type
- application/json
{
"instances": [
{
"messages": [
{
"role": "user",
"content": [
{
"type": "text",
"text": "Analyze this workplace safety video and identify potential hazards."
},
{
"type": "video_url",
"video_url": {
"url": "UklGRnoGAABXQVZFZm10...",
"fps": 2.0,
"max_frames": 16
}
}
]
}
],
"parameters": {
"max_new_tokens": 150,
"temperature": 0.6,
"do_sample": true,
"return_history": true
}
}
]
}
{
"instances": [
{
"messages": [
{
"role": "user",
"content": [
{
"type": "text",
"text": "Analyze this workplace safety video and identify potential hazards."
},
{
"type": "video_url",
"video_url": {
"url": "https://your-bucket.s3.amazonaws.com/videos/safety.mp4?AWSAccessKeyId=ASIA...&Signature=...",
"fps": 2.0,
"max_frames": 16
}
}
]
}
],
"parameters": {
"max_new_tokens": 150,
"temperature": 0.6,
"do_sample": true,
"return_history": true
}
}
]
}
Input data descriptions
The following table describes supported input data fields for real-time inference and batch transform.
Field name | Description | Constraints | Required |
|---|---|---|---|
max_new_tokens | Maximum tokens to generate in response. Default value is 50. | 1-1024 | No |
do_sample | Enable sampling. Default value is true. | true/false | No |
temperature | Controls randomness in generation. Lower values produce more deterministic outputs, higher values produce more creative outputs. Set do_sample=False to disable sampling (equivalent to temperature=0). Default: 0.6 | 0.0-2.0 | No |
fps | Video frame rate for processing. Lower values (0.5-1.0) recommended for longer videos. Default value is 1.0. | 0.0-10.0 | No |
max_num_frames | Maximum number of frames to sample from the input video. Used together with fps to calculate the frame budget (fps × video duration, capped at this value). Default: 128 | 1-2048 | No |
top_p | Nucleus sampling threshold. Only tokens with cumulative probability above this value are considered. Only applies when do_sample=True. | 0.1–1.0 | No |
top_k | Limits token selection to the top-k most probable tokens at each step. Only applies when do_sample=True. Omit this parameter to disable top-k filtering. Default: 20 | 1–100 | No |
return_history | Return conversation history. Default value is true. | true/false | No |
chat_history | Chat history from previous prompts. Available in return_history field. Default value is false. | true/false | No |
Resources
Vendor resources
Support
Vendor support
Technical Support: Please contact us if you need support.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
Woven Solutions is dedicated to delivering innovative cloud-native solutions to the federal government, specializing in Cloud Engineering, Software Development, Cybersecurity, and DevOps. Our deep expertise enables us to provide tailored, cutting-edge solutions addressing the unique challenges of federal clients.
Includes global market data on garments, fabrics, yarns, and apparel items. Covers production, trade, and pricing data from 2012 to 2035. Data is available for all countries and major regions worldwide. Approx. 130 products are included based on HS code classification.
This dataset enables data-driven decisions by providing actionable insights into global trade, pricing, production, and market trends. It supports strategic planning, market assessment, and forecasting across industries and regions.
Cloud Security Architecture Assessment
NCC Group can help to assure, secure, and create architectural solutions. Our modular approach enables us to provide a customised solution tailored to your security needs. We will conduct workshops to collect, analyze, summarize, and understand your solution requirements. We will evaluate your assets, security controls, review your configurations and give actionable recommendations and next steps to improve cyber security maturity.
We:
-Review existing solutions
-Contribute to design teams to add resiliency and security-by-design to cloud and hybrid environments
-Provide a holistic service with solutions to mitigate the risks associated with cloud-based technologies.
NCC Group collaborates with you to develop architectures which are resilient against exploitation, limit the impact of any compromise and facilitate continuous development and improvement.
DryRun Security helps software teams ship secure code with confidence. Our AI powered platform analyzes code changes in real time, detecting risky behaviors that traditional SAST tools miss. We focus on developer first workflows, empowering engineers to catch security issues as they are introduced, while giving AppSec teams the visibility and context they need to manage risk at scale.
Customer reviews
No customer reviews yet
Be the first to review this product . We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.