The professional standard for AI upscaling — 10 specialized models, built-in face enhancement, and generative detail recovery. Trusted by Apple, Netflix, and NASA. From $0.08/image.
Image and video upscaling endpoints on Runflow.
10 specialized AI models for image enhancement
AI video upscaling and enhancement
Every model on Runflow is scored per vertical — not generic leaderboard numbers. Scores come from domain-expert evaluation on real production prompts.
| Use Case | Assessment | Notes |
|---|---|---|
| Photo Upscale | Pending | Photo upscaling quality |
| AI Art Upscale | Pending | AI-generated image upscaling |
| Face Enhancement | Pending | Face restoration and enhancement |
| Document/Text | Pending | Scanned text clarity |
| Output resolution | Price per image |
|---|---|
| Up to 24 MP | $0.08 |
| Up to 48 MP | $0.16 |
| Up to 96 MP | $0.32 |
| Up to 512 MP | $1.36 |
For a single image, pricing is tiered by output megapixels. Cold start time is not billed.
Full parameter reference for the topaz-upscale-image model endpoint.
| Parameter | Type | Description |
|---|---|---|
| image_url | string | URL of the image to upscale. Accepts JPEG, PNG, WebP, GIF, AVIF. |
| model | string | AI model for enhancement. Options: Standard V2 (default), Low Resolution V2, CGI, High Fidelity V2, Text Refine, Recovery, Redefine, Recovery V2, Standard MAX, Wonder. |
| upscale_factor | number | Upscale factor from 1x to 4x (continuous float). Default: 2. |
| face_enhancement | boolean | Enable automatic face detection and enhancement. Default: true. |
| face_enhancement_strength | number | Face enhancement intensity (0–1). Default: 0.8. |
| face_enhancement_creativity | number | Face generation creativity (0–1). Higher values hallucinate more detail. Default: 0. |
| subject_detection | string | Target enhancement region. Options: All (default), Foreground, Background. |
| sharpen | number | Sharpening level (0–1). Works with Standard V2, Low Resolution V2, CGI, High Fidelity V2, Text Refine, Redefine. |
| denoise | number | Noise reduction level (0–1). Works with Standard V2, Low Resolution V2, CGI, High Fidelity V2, Text Refine, Redefine. |
| fix_compression | number | JPEG artifact removal (0–1). Works with Standard V2, Low Resolution V2, High Fidelity V2, Text Refine. |
| detail | number | Detail recovery level (0–1). Recovery V2 model only. |
| creativity | integer | Generative upscaling creativity (1–6). Redefine model only. |
| texture | integer | Texture detail level (1–5). Redefine model only. |
| prompt | string | Text prompt to guide generative upscaling (max 1024 chars). Redefine model only. |
| autoprompt | boolean | Auto-generate a prompt for generative upscaling. Redefine model only. |
| output_format | string | Output image format. Options: "jpeg", "png". Default: "jpeg". |
| crop_to_fill | boolean | Crop the output image to fill target dimensions. Default: false. |
Flux AI models are available on many platforms. Here's what makes the Runflow image generation API different.
Choose the right model for your content: Standard V2 for photos, CGI for digital art, Text Refine for documents, Recovery V2 for damaged images, Redefine for generative upscaling, Wonder for all-in-one enhancement.
Automatic face detection and enhancement with adjustable strength (0–1) and creativity (0–1). Enabled by default — no separate face restoration step needed.
Prompt-guided creative detail generation with creativity (1–6) and texture (1–5) controls. Rebuilds structure with diffusion-like enhancement — not just interpolation.
Denoise, sharpen, fix compression artifacts, and upscale in a single API call. No need to chain multiple tools or models.
Target enhancement to All, Foreground only, or Background only — apply different treatment to subjects vs. backgrounds in one pass.
Same API, different strengths. Switch models with one parameter change.
Black Forest Labs
FLUX.1 [dev] is a 12B-parameter text-to-image model by Black Forest Labs. Open-w…
ByteDance
Seedream 4.5 by ByteDance — unified generation and editing in one model. Strong…
ByteDance
Seedream 4 by ByteDance — unified generation and editing model with photorealism…
Nano Banana 2 by Google — fast image generation across 4 resolution tiers (0.5K–…
OpenAI
GPT-Image 1.5 by OpenAI — high-fidelity images with strong prompt adherence. Thr…
Nano Banana Pro by Google — premium image generation with sharp detail and natur…
Get API access in minutes. No GPU setup, no infrastructure to manage. Pair with Sentinel to control the quality of every AI-generated image you ship.