Step 3.7 Flash (free)
Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model. It pairs a 196B-parameter language backbone with a vision encoder for native image and video understanding, activating roughly 11B parameters per token. The model supports a 256K context window and exposes selectable reasoning levels (high/medium/low), letting callers trade off speed, cost, and depth of reasoning. Designed for coding, agentic workflows, structured outputs, and long-context productivity tasks.
Offers
Same model, different listings. First-party, router, and aggregator rows stay separate so you can see the spread instead of a single blended number. Showing free/$0 offers. Hide them.
| Source | Provider | Input | Output | Cache read | Context |
|---|---|---|---|---|---|
| models.devcheapeststepfun/step-3.7-flash:free | Kilo Gateway | $0 | $0 | $0 | 262K |
| models.devstepfun/step-3.7-flash-free | ZenMux | $0 | $0 | $0 | 256K |
| models.devstep-3.7-flash:free | UnoRouter | $0 | $0 | $0 | 256K |
Capabilities
- Reasoning
- Tool calling
- Attachments
- Open weights
- In: text
- In: image
- In: video
- Out: text
Scores
Sparse coverage on purpose. Missing a score means we have not linked one yet, not that the model is unranked. Full boards on /benchmarks.
- AA intelligence30.3
- AA coding39.6
- AA agentic21.5
Sources: Arena AI, Artificial Analysis. See /benchmarks.