Sold by: Superb AI
Deployed on AWS
Free Trial
"ZERO" is an industrial Vision Foundation Model (VFM), ready for immediate deployment without the need for data labeling or model retraining. Superb AI's "ZERO" leverages zero shot and open world technologies to instantly detect and pinpoint objects in images and videos using simple text, box prompts.
Overview
Traditional Vision AI demands extensive data labeling and repetitive model retraining, a labor-intensive process that consumes significant time, cost, and specialized expertise. Superb AI's "ZERO" brings a paradigm shift as an industrial-specialized Vision Foundation Model (VFM). Leveraging Open World Visual Grounding technology, ZERO comprehends novel concepts without prior training. This zero-shot capability empowers instant AI adoption for new tasks and flexible, on-the-fly changes to detection targets, eliminating the need for additional training.
A key advantage of ZERO is its Multi-Prompt capability. This feature enables immediate AI model application in real-world scenarios using intuitive prompts like text, image boxes, no separate model training required. Instead of time-consuming retraining, you simply describe your target in text or provide an example image, and ZERO adapts instantly. This dramatically cuts the time and cost of AI solution development, making AI adoption faster and more accessible.
Furthermore, ZERO is engineered for high efficiency, boasting a lightweight 622M parameters and high-performance processing at 1.03 TFLOPS. This optimized design ensures seamless operation across virtually any application or industry, from cloud to secure on premise environments and remote edge devices. Its efficiency significantly reduces the need for heavy computing hardware investments, making advanced AI more attainable for diverse industrial deployments.
Highlights
- Zero shot Deployment: Instantly detect untrained objects without complex data collection, labeling, or model retraining. Adapt immediately to new products, defect types, or environment changes, dramatically cutting development time and costs.
- Flexible Multi Prompt Input: Deploy and operate AI instantly by simply describing your target object in text or providing an example image. ZERO supports diverse input prompts for intuitive interaction.
- Industrial Specialized VFM: Trained on invaluable, real-world data from dozens of industrial sectors including manufacturing, logistics, and retail. ZERO delivers high performance and immediate usability across complex industrial domains.
Details
Sold by
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
Free trial
Try this product free for 5 days according to the free trial terms set by the vendor.
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.g4dn.xlarge Inference (Batch) Recommended | Model inference on the ml.g4dn.xlarge instance type, batch mode | $5.00 |
ml.g4dn.xlarge Inference (Real-Time) Recommended | Model inference on the ml.g4dn.xlarge instance type, real-time mode | $5.00 |
Dimensions summary
You pay by the hour for model inference on the ml.g4dn.xlarge GPU instance. Billing is usage-based, so charges accrue per host hour the instance runs. Two options set the inference method. Batch mode processes groups of data at once. Real-time mode returns results on demand as requests arrive. Both run on the same instance type, so the difference is how the model handles requests, not the underlying hardware. Pick the mode that fits your workload, or run both. Costs scale with how many hours each mode operates.
Top-of-mind questions for buyers
You pay for each hour the ml.g4dn.xlarge GPU instance runs your inference workload. This instance uses a single GPU. One host hour means one hour of that instance running, billed regardless of how many detections you process during that hour.
Charges accrue while the instance runs and stop when it stops. Real-time mode keeps the instance running to answer requests on demand, so hours accrue continuously. Batch mode runs the instance for the length of each processing job, so hours accrue only during active jobs.
Each mode meters host hours independently on the same instance type. If you run both, your bill adds the hours from batch jobs to the hours from real-time serving. Neither mode requires the other, so you can run one, the other, or both.
superb-ai.com
Helpful?
Vendor refund policy
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
Enables object detection by text and vision based prompt in single image
Additional details
Inputs
- Summary
Parameters
search_image Type: String Required: Yes Description: The source image for object detection. This can be a URL or a Base64 encoded string.
queries Type: List[Dict] Required: Yes Description: A list of query objects that define what to search for in the source search_image. At least one query must be provided.
-
prompt_image Type: String Required: Conditional Description: An image used for semantic queries. This is required when you want to use a part of an image (defined by a bounding box) as the search prompt. Can be an empty string "", URL or a Base64 encoded string.
-
prompts Type: List[Dict] Required: Yes Description: A list of prompt objects, each defining a specific target to detect. The prompts Object Each object within the prompts list specifies the actual search criteria. This is where you define whether the query is text-based, semantic (visual), or a combination.
-
- text Type: String Required: Conditional Description: A text description of the object to detect (e.g., "person", "red car"). Required for text-based detection. Can be an empty string "" if a box is provided for a semantic query.
-
- box Type: List[Float] Required: Conditional Description: A list of four floats representing a bounding box [x, y, w, h] that crops a region from the prompt_image. This cropped region is used as a visual prompt for detection. Required for semantic (visual) search. Provide an empty list [] for text-based search.
-
- box_threshold Type: Float Required: Yes Description: A threshold between 0.0 and 1.0. Detection boxes with a score below this value will be filtered out.
-
- multimodal_threshold Type: Float Required: Yes Description: A threshold between 0.0 and 1.0. Each corresponding threshold will be applied. to the individual prompt. Try to use this threshold for the product rather than box_threshold.
-
- Input MIME type
- application/json, application/jsonlines
{"search_image": "https://www.californiastrawberries.com/wp-content/uploads/2021/05/Rainbow-Fruit-Salad-1024.jpg", \
"queries": [{"prompt_image": "", \
"prompts": [{"text": "strawberry", "box": [], "box_threshold": 0.1, "multimodal_threshold": 0.22}]}]}
{"search_image": "https://www.californiastrawberries.com/wp-content/uploads/2021/05/Rainbow-Fruit-Salad-1024.jpg", \
"queries": [{"prompt_image": "", \
"prompts": [{"text": "strawberry", "box": [], "box_threshold": 0.1, "multimodal_threshold": 0.22}]}]}
Support
Vendor 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
Kriv AI builds a continuous agent quality + safety testing harness on Amazon Bedrock. Scope: regression suite (100–1,000 golden test cases per agent), jailbreak corpus aligned to OWASP Top 10 for LLM Applications (2024/2025 — LLM01–LLM10), bias probes (disparate-impact testing via AIR / four-fifths / statistical parity / equalized odds / calibration / Cohen's d + protected-class proxy detection via SageMaker Clarify), output validation (JSON schema + toxicity + hallucination + grounding), incident playback (time-travel debugging via OpenTelemetry + X-Ray), CI/CD integration (CodePipeline / GitHub Actions / GitLab CI / Jenkins), eval dashboards (CloudWatch + QuickSight). Integrates with Bedrock Model Evaluations + AgentCore Evaluations (GA March 2026). Three tiers: $50K / $85K / $125K + $20K Extra Agent. Optional $8K–$15K/month Retainer. AWS Select + Anthropic CPN.
Zero-day Malware Protection for Your Cloud and Hybrid Workloads
Zero-day Malware Protection for Your Cloud and Hybrid Workloads
NZTA is a cloud-based SaaS (Software as a Service) application that provides fully-managed zero-trust authentication and access control for an organization's application infrastructure.2 Nic M5 supported version.
Zero Trust Secure Access enables you to provide purpose-built connections between your applications in AWS and your employees.
This is a repackaged open source software product wherein additional charges apply for a pre-hardened AMI with automated startup. Self-hosted Cloudflare Tunnel client for exposing local services securely. No port forwarding, no firewall rules.
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.