Topaz Upscale Image Creative: AI 图像编辑器

Upscale images with Topaz Creative. Add plausible new fine detail up to 4x resolution with Bloom 2/Bloom/Bloom Realism. Try free — pay per generation.

交互式模型控制

Upscale images with Topaz Creative. Add plausible new fine detail up to 4x resolution with Bloom 2/Bloom/Bloom Realism. Try free — pay per generation.

📝

概览

关于此模型

Topaz Upscale Image Creative enlarges an image up to 4x while inventing new, plausible fine detail — hair strands, fabric weave, skin texture — rather than just smoothing pixels. It runs on Topaz's Bloom 2, Bloom, and Bloom Realism models, with optional autoprompting and adjustable creativity so you can dial in how much new detail the model is allowed to imagine versus how closely it should stay to the source. Works well on soft phone photos, old scans, and AI-generated images that need a sharper, more detailed finish. See also Topaz Upscale Image Generative for a more faithful reconstruction-focused alternative, or Topaz Upscale Image Precision when you need to preserve the original look exactly.

1Creative restoration: Turn a low-res portrait into a sharp, detailed 4x enlargement for print.
2AI art finishing: Add fine texture and detail to AI-generated images before final delivery.
3Marketing assets: Upscale product renders and social graphics without a visible loss of sharpness.
4Archival photos: Breathe new detail into old, blurry, or low-resolution family photos.
💰

价格与价值

成本分析

muapiapp$0.08+ per started 2 megapixels of output (plus margin), billed dynamically

按用量付费,无订阅。费用根据请求时的实际输出分辨率/时长计算。

Fal.ai暂不可用

muapiapp 的上游基础设施合作方目前未将其作为独立的公共 API 提供。

Replicate暂不可用

目前未在 Replicate 上架。

** 竞品价格根据相似模型架构和使用层级估算。

⚙️

技术详情

配置参数

图像 URLstring

Input image to upscale.

默认值https://d3adwkbyhxyrtq.cloudfront.net/webassets/imagemodels/topaz-upscaler-in.jpg
模型枚举(3 个选项)

Topaz creative upscale model variant.

默认值Bloom 2
放大倍数number

Factor to upscale the image by.

默认值2
Autopromptboolean

Automatically generate a creative-detail prompt.

默认值-
创意度int

Creative detail strength (1-9).

默认值-
Color Preservationboolean

Preserve original color palette.

默认值-
Crop To Fillboolean

Crop the output to exactly fill the target aspect ratio.

默认值false
输出格式枚举(2 个选项)

Output image file format.

默认值jpeg
📖

实施指南

开发者文档

How to Use Topaz Upscale Image Creative

  1. Provide an image: Set image_url to any publicly accessible image URL.

  2. Pick a model: Choose Bloom 2 (default, latest), Bloom, or Bloom Realism depending on how photoreal you want the added detail.

  3. Tune creativity: Set creativity (1-9) to control how much new detail the model invents, and color_preservation to keep the original palette.

  4. Set the scale: upscale_factor accepts 1-4x. Choose output_format (jpeg or png).

  5. Submit and poll: POST to /api/v1/topaz-upscale-image-creative, then poll /api/v1/predictions/{request_id} (or use webhook_url) until status is completed.

curl -X POST https://api.muapi.ai/api/v1/topaz-upscale-image-creative \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "image_url": "https://example.com/photo.jpg",
    "model": "Bloom 2",
    "upscale_factor": 2,
    "creativity": 5
  }'
❓

常见问题

常见问答

What does Topaz Upscale Image Creative do?

It enlarges an image up to 4x while generating new, plausible fine detail using Topaz's Bloom family of creative upscale models.

How is Creative different from the Generative or Precision Topaz upscalers?

Creative leans furthest toward inventing new detail for a more stylized, sharpened result. Generative reconstructs realistic detail (faces, textures) more conservatively, and Precision stays closest to the original pixels.

What image formats and sizes are supported?

Any standard JPEG/PNG/WebP URL works as input. Output can be requested as `jpeg` or `png` via `output_format`.

How is pricing calculated?

Cost is based on the actual output megapixels after `upscale_factor` is applied, billed per started 2-megapixel block.

Can I control how much new detail is added?

Yes — the `creativity` parameter (1-9) controls how much new detail the model invents, and `color_preservation` keeps the original color palette intact.

Is there a webhook for completion?

Yes, pass `webhook_url` in the request body or as a `?webhook=` query parameter and muapi will POST the final result when processing completes.