ShieldstralMistral AI
|
||||||
Related Products
|
||||||
About
Preamble's AI Safety and Security Platform is an integrated solution designed to streamline and enhance the management of AI systems within an organization. It offers a centralized hub for managing people, overseeing diverse data labeling projects, providing clear guidelines for consistent data labeling, and tracking all labels and datasets.
The platform also facilitates the evaluating of custom models and serves as a comprehensive center for AI safety and security testing and policy deployment. From real-time engagement with AI models to rigorous policy testing, the platform combines these multifaceted components to ensure alignment with organizational values, ethical principles, and compliance standards. Whether it's managing individual roles, conducting adversarial testing, or deploying safety controls, Preamble's platform offers a cohesive and user-friendly environment that addresses the complex and evolving needs of AI safety and security.
|
About
Shieldstral is a 3B open-weights, policy-adaptive multimodal safety classifier designed to evaluate text, images, and text-plus-image content using policies defined at inference time. Instead of relying on a fixed taxonomy of harm categories, it frames moderation as a binary question-answering task: users provide an instruction describing the evaluation context and strictness, a yes-or-no safety question, and the content to judge. The model reads the “yes” and “no” logits and converts them into a continuous, calibrated safety score, allowing applications to threshold or rank results by confidence rather than depend on a single discrete label. This formulation unifies prompt classification, response moderation, refusal detection, toxicity detection, and multimodal safety in one interface, while letting teams adapt policies without retraining the model. Shieldstral can evaluate prompts, responses, prompt-response pairs, images, and images with accompanying text.
|
|||||
Platforms Supported
Windows
Mac
Linux
Cloud
On-Premises
iPhone
iPad
Android
Chromebook
|
Platforms Supported
Windows
Mac
Linux
Cloud
On-Premises
iPhone
iPad
Android
Chromebook
|
|||||
Audience
Any company using generative AI and LLMs.
|
Audience
AI platform teams that need customizable, multimodal content moderation without retraining a separate safety model for every policy
|
|||||
Support
Phone Support
24/7 Live Support
Online
|
Support
Phone Support
24/7 Live Support
Online
|
|||||
API
Offers API
|
API
Offers API
|
|||||
Screenshots and Videos |
Screenshots and Videos |
|||||
Pricing
$100/month/user
Custom prices available for enterprises.
Free Version
Free Trial
|
Pricing
No information available.
Free Version
Free Trial
|
|||||
Reviews/
|
Reviews/
|
|||||
Training
Documentation
Webinars
Live Online
In Person
|
Training
Documentation
Webinars
Live Online
In Person
|
|||||
Company InformationPreamble
Founded: 2021
United States
www.preamble.com
|
Company InformationMistral AI
Founded: 2023
France
mistral.ai/news/shieldstral/
|
|||||
Alternatives |
Alternatives |
|||||
|
|
|
|||||
|
|
||||||
|
|
|
|||||
Categories |
Categories |
|||||
Content Moderation Features
Artificial Intelligence
Audio Moderation
Brand Moderation
Comment Moderation
Customizable Filters
Image Moderation
Moderation by Humans
Reporting / Analytics
Social Media Moderation
User-Generated Content (UGC) Moderation
Video Moderation
|
||||||
Integrations
ChatGPT
Claude
Cohere
IBM watsonx.data
Llama 2
Mistral AI
OpenAI
Reddit
|
Integrations
ChatGPT
Claude
Cohere
IBM watsonx.data
Llama 2
Mistral AI
OpenAI
Reddit
|
|||||
|
|
|