SAM 3

SAM 3 is a unified foundation model for promptable segmentation in images and videos.

By Meta

Pricing: $0.005 per image

Overview

Built by Meta, Segment Anything Model 3 is a image editing model. SAM 3 is a unified foundation model for promptable segmentation in images and videos. It can detect, segment, and track objects using text or visual prompts such as points, boxes, and masks. Users can expect precision edits that preserve the original's integrity in production. Automatic provider failover keeps your pipeline running when others go down.

Examples

• Demo of text-prompted segmentation. Input is an editorial fashion portrait generated by openai/gpt-image-2; sam-3 isolates the glasses with the prompt 'glasses' and overlays the mask on the original.

Frequently asked questions

What is Segment Anything Model 3?

SAM 3 is a unified foundation model for promptable segmentation in images and videos. It can detect, segment, and track objects using text or visual prompts such as points, boxes, and masks.

How much does Segment Anything Model 3 cost on Runflow?

Contact us for current pricing on Segment Anything Model 3. Runflow offers predictable, usage-based pricing with no GPU management fees. Request a quote at runflow.io/pipeline-audit.

What's the latency of Segment Anything Model 3?

Latency for Segment Anything Model 3 varies based on input parameters and current load. Runflow optimizes inference with multi-region routing to minimize response times. Cold starts (5-10s) are not billed.

Can I use Segment Anything Model 3 commercially?

Yes. Segment Anything Model 3 is available under a commercial license, meaning you can use it in production applications, SaaS products, and client work. Check the model's specific license terms for details.

Do I need to manage GPUs?

No. Runflow handles all infrastructure, including GPU provisioning, scaling, and failover. You call our API and receive results — no DevOps required.

What other models does Runflow support?

Runflow supports 736 models across image generation, video, audio, 3D, and more. Browse the full catalog at runflow.io/models or contact us for recommendations based on your use case.

Related models

• Logo Inpaint, Place a logo onto a target image, with optional mask to constrain placement. When the mask is omitted, the workflow picks a placement automatically.
• Object Removal (Prompt), Remove requested objects from a photo using a short text instruction. Flux-fill restores the background without requiring a mask.
• 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.
• 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.

Discoverable surfaces

• Dispatch endpoint: POST https://api.runflow.io/v1/models/sam-3/image/runs
• Per-model spec (markdown): https://app.runflow.io/models/sam-3/image/llms.txt
• Docs page: https://docs.runflow.io/models/sam-3/image
• Public OpenAPI spec: https://docs.runflow.io/api/openapi.public.json
• Agent skill (start here): https://www.runflow.io/.well-known/agent-skills/runflow/SKILL.md