Stable Diffusion is an open-source image generation model developed by Stability AI. It enables users to generate images from text prompts such as a description of an object, background, or style.
The model is powered by Baseten and is governed by the CreativeML Open RAIL M License. This allows users to generate images with just a descriptive phrase, such as 'A lion wearing a cowboy hat'.
The model then generates the corresponding image, with the example output provided in the text including a cyberpunk portrait, a lion wearing a cowboy hat, an NFL player scoring a touchdown, and a digital art of a tranquil library on a spaceship.
This tool provides a convenient and powerful way for users to create images from text prompts.
High-Performance Inference: Dedicated infrastructure for serving open-source, custom, and fine-tuned AI models optimized for high-scale workloads.
Pre-Optimized Model APIs: Instant access to the fastest models in production for testing new workloads and prototyping products.
Multi-Cloud Deployment Options: Ability to scale workloads across any cloud provider with options for self-hosted and hybrid deployments.
Rapid Image Generation and Transcription: Supports ultra-fast image generation and optimized transcription services with high accuracy and low latency.
Forward Deployed Engineering Support: Access to engineering expertise for building, optimizing, and scaling models from prototype to production.
Rapid image generation for creating high-quality visuals in applications like presentations and social media content.
Optimized transcription services for accurate and efficient speech-to-text conversion and speaker diarization.
Real-time text-to-speech capabilities for applications such as voice agents and AI phone calls.
High-performance large language model (LLM) runtimes for coding assistants and chat systems.
Deployment of custom and fine-tuned AI models for specific business needs, ensuring cost-effective and scalable solutions.
Provides high-performance inference for custom and open-source AI models, enabling efficient deployment at scale.
Offers rapid model testing and prototyping with pre-optimized APIs, facilitating quick iterations and product development.
Ensures low latency and high throughput for applications like transcription, image generation, and text-to-speech, enhancing user experience.
Supports seamless integration with existing workflows through a developer-friendly environment, reducing the need for extensive engineering resources.
Delivers cost-effective solutions with flexible pricing models, allowing businesses to scale without financial strain.
Basic Plan: $0 per month, pay as you go; includes dedicated deployments, model APIs, training, fast cold starts, and email/in-app chat support.
Pro Plan: Includes everything in Basic plus priority access to high-demand GPUs, dedicated compute, higher model API rate limits, and hands-on engineering expertise; volume discounts available.
Enterprise Plan: Includes everything in Pro plus custom SLAs, self-host deployments, on-demand flex compute, and advanced security; volume discounts available.
Model API Pricing for DeepSeek-V4-Flash-0731:
Input: $0.13 per 1M tokens
Cache Input: $0.028 per 1M tokens
Output: $0.26 per 1M tokens
Dedicated Deployments: Prices for GPU instances range from $0.01052 per minute for T4 to $0.16633 per minute for B200.
