Stable Diffusion 1.5

Stable Diffusion v1.5 is a diffusion model for image generation based on textual prompts. The model was initialized with the weights of the previous version, Stable Diffusion v1.2, and subsequently fine-tuned over 595,000 steps at a resolution of 512×512 using the "laion-aesthetics v2 5+" dataset with 10%. The dataset contains adult content without additional deduplication measures.
The model uses the CLIP ViT-L/14 text encoder for text encoding and the U-Net architecture for image generation. It supports image generation at 512×512 pixels and image modification via textual prompts. During training, 32 x 8 x A100 GPUs were used with the AdamW optimizer at a learning rate of 0.0001 and a batch size of 2048.

Announce Date: 01.10.2022
Parameters: 1B
Context: 77
VRAM requirements: 0.5 GB using 4 bits quantization
Developer: RunwayML
License: CreativeML OpenRail-M

Use our pre-built public endpoints to test inference and explore Stable Diffusion 1.5 capabilities.

Rent your own physically dedicated instance with hourly or long-term monthly billing.

We recommend deploying private instances in the following scenarios:

• maximize endpoint performance,
• enable full context for long sequences,
• ensure top-tier security for data processing in an isolated, dedicated environment,
• use custom weights, such as fine-tuned models or LoRA adapters.

Contact our dedicated neural networks support team at nn@immers.cloud or send your request to the sales department at sale@immers.cloud.