Smart Segmentation
Segment a requested subject from an image using the SAM3 VLM workflow. Returns one or more mask/segmentation outputs from a prompt-guided image analysis pass.
By Runflow
Pricing: $0.12 per request
Overview
Smart Segmentation extracts a prompt-selected subject or region from an input image and returns a mask through the standard Runflow output contract. Send an image URL and a concrete prompt such as "left shoe", "mug handle", or "character head" to target the region you need.
Key capabilities
- **$0.12 per successful image**, fixed per-run pricing for prompt-guided segmentation masks.
- **Multi-region failover**, Runflow runs optimized inference across multiple regions. If one region has an outage, traffic routes to another automatically.
- **Prompt-guided selection**, useful when a scene contains multiple objects and you need a specific part instead of the whole foreground.
- **Simple API shape**, production calls use `image_url` and `prompt`; internal filename handling is inferred automatically from the URL.
Technical specifications
| Spec | Details |
|---|---|
| Input | Image URL plus text prompt |
| Output | Image mask in `outputs[]` |
| Category | Image to image |
| Pricing | $0.12 per successful run |
Examples
- CLI smoke test from completed BetterInfra job
- Generated through draft Runflow dispatch during CLI audit.
Frequently asked questions
- What is Smart Segmentation?
- Smart Segmentation is a Runflow image model that creates a mask for a prompt-selected subject or region in an image. It is designed for cases where you need a specific object part, product region, or character component instead of a full foreground cutout.
- How much does Smart Segmentation cost on Runflow?
- Smart Segmentation costs $0.12 per successful image run on Runflow. Failed or canceled runs are not intended to be used as billable production outputs.
- What inputs does Smart Segmentation need?
- Smart Segmentation needs an image_url and a prompt. The prompt should be concrete, for example "left shoe", "mug handle", "jacket collar", or "character head". Runflow infers the internal input filename from the image URL automatically.
- What does Smart Segmentation return?
- Smart Segmentation returns a mask image in the standard Runflow outputs array. You can inspect, composite, or post-process that mask in your own pipeline.
- Can I use Smart Segmentation commercially?
- Smart Segmentation is available through Runflow for commercial API use, subject to your Runflow account terms and acceptable-use policies.
- Do I need to manage GPUs for Smart Segmentation?
- No. Smart Segmentation runs through Runflow infrastructure, so you call the API and Runflow handles dispatch, inference, retries, and regional availability.
Related models
- Object Removal (Prompt), Remove requested objects from a photo using a short text instruction. Flux-fill restores the background without requiring a mask.
- Background Replace, Swap the background of any product photo for a new scene defined by a reference image and a short prompt. The model preserves the foreground subject while matching camera angle and lighting to the new background.
- Reference Inpaint, Inpaint a region of a target photo using a reference image as the visual guide. Provide the target, a black-and-white mask of the area to fill, and a reference image. The model paints the masked region with content matching the reference.
- Product Isolation, Isolate the main product from a busy scene onto a clean background, using an optional text prompt to disambiguate the subject. Built for studio-grade product photography at scale.