Step-1V
stepfun/step-1vVision-first model from a lab that trains multimodality natively rather than bolting it on.
Model overview
Context window
32K
tokens
Input price
—
quoted on request
Output price
—
quoted on request
Weekly volume
—
tokens / week
Open weights
No
API only
Capabilities
2
of 11 tags
Description
Step-1V is a vision model from StepFun. Webparam does not stock this model yet and can source it on request — tell us what you need it for and we will come back with availability and a price. StepFun’s stated focus is multimodal foundation models. The weights are not published, so it is API-only.
Capabilities
- Vision input
- Accepts images alongside text in the prompt.
- Streaming
- Emits tokens as they are generated, so answers appear progressively rather than all at once.
Strengths & weaknesses
Strengths
- Natively multimodal training
- Good spatial reasoning
- Stable outputs
Weaknesses
- 32K context
- Slower ecosystem adoption
Pricing
| Rate | Price | Unit |
|---|---|---|
| Input | — quoted on request | — |
| Output | — quoted on request | — |
This model is not provisioned yet; the rate is quoted on request.
Context window
32Ktokens
covers prompt and response together, so a long input leaves less room for the answer.
Supported features
| Feature | Support |
|---|---|
| Tool calling | Not supported |
| JSON mode | Not supported |
| Streaming | Supported |
| Vision input | Supported |
| Audio input | Not supported |
| Long context | Not supported |
| Open weights | Not supported |
| Reasoning | Not supported |
| Fine-tunable | Not supported |
| Batch | Not supported |
| Caching | Not supported |
Example use cases
Visual reasoning
Scene understanding
Multimodal research
Comparison
Compare against anything →| Attribute | Step-1V | Step-2 | Hunyuan Vision |
|---|---|---|---|
| Context window | 32K | 128K | 128K |
| Input price | R0.00 | R0.00 | R0.00 |
| Output price | — not charged | — not charged | — not charged |
| Tool calling | Not supported | Supported | Not supported |
| JSON mode | Not supported | Supported | Not supported |
| Vision input | Supported | Not supported | Supported |
| Open weights | Not supported | Not supported | Not supported |
| Prompt caching | Not supported | Not supported | Not supported |
Best value in each row is highlighted. Blank cells are facts the supplier does not publish, not zeroes.
Provider information
View StepFun →StepFun · 阶跃星辰 · Shanghai, China · est. 2023
StepFun builds trillion-parameter-class multimodal systems, its Step series spans text and vision with an emphasis on unified multimodal training rather than bolted-on adapters. A quieter lab than its peers, it is consistently cited in multimodal research.
Recent releases
Documentation
Frequently asked questions
Input is billed at R0.00 per 1M tokens, and there is no separate output charge. Every figure on this page is an illustrative placeholder.
No. Step-1V is served through its provider's API only; the weights are not published, so self-hosting is not an option.
Prompt and response share one window: Step-1V accepts up to 32K tokens in total, so a long input leaves less room for the answer. The median across vision models is 128K. Recall is strongest well inside the limit. Test at your real prompt depth before relying on the ceiling.
Step-2 is the closest neighbour in the catalogue, with 128K of context at R0.00 per 1M tokens in. Put both in a comparison before wiring a fallback chain, so the trade-off is a decision rather than a default.