Cost per second for a slow inference
$ 0.001000The speed depends on the GPU on which the model is run.
To get faster inferences please contact our sales team.
Cost per second for a fast inference
$ 0.001500For enterprise pricing and custom weights or models
contact our sales team.
Realvis Lightning SDXL
The Realvis Lightning SDXL model is an advanced solution designed for swift and efficient generation of high-quality, photorealistic images. This model, an enhanced iteration of the Realvis XL, excels in delivering exceptional photorealistic images with remarkable speed. Its standout feature lies in its capacity to produce top-tier 1024px images rapidly, catering to applications that necessitate quick image generation.
To maximize the Realvis Lightning SDXL model's performance, it is crucial to ensure compatibility with the DPM++ SDE Karras / DPM++ SDE sampler. Optimal utilization involves employing 4-6 sampling steps and a CFG Scale of 1-2. These specific settings are key to achieving an effective and efficient image generation process, enabling users to fully leverage the capabilities of this sophisticated model.
