Step-1o Turbo Vision
Step-1o Turbo Vision is a hosted visual understanding model. It accepts text, images and video, and returns text. Its 32K context is smaller than the current Step 3.7 and Step 5 models.
Add information that the image cannot carry
The visual guide states that preprocessing removes image metadata. File names, camera information and location metadata are therefore not evidence that the model can read from the image.StepFun visual understanding models
Use it for a bounded visual task
Practical guidance. This model can fit a short caption, image question or video summary. Add relevant file context as text when it matters. A large collection of images needs a plan for context use even when it stays below the file-size limit.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | step-1o-turbo-vision |
| Context window | 32,000 tokens Provider-listed 32K context. StepFun visual understanding models |
| Request size | The public catalog lists up to 60 images, 20 MB total image data, and MP4 video below 128 MB.StepFun public model catalog |
| Migration role | StepFun names this model as the replacement for several retired Step 1 and Step 2 APIs.StepFun model retirement and migration |
Related models
- Step 5 Preview
StepFun · Step 5
A multimodal mixture-of-experts model with 600B total and 27B active parameters for text, image and video tasks.
Launch date not confirmedPreviewAPI · Downloadable · Provider product - Step 3.7 Flash
StepFun · Step 3
A visual reasoning model with a 196B language backbone and 1.8B vision encoder for images, video, code and tool calls.
Launch date not confirmedAPI · Downloadable