The Orchestrator › Models › StepFun
StepFun: Step 3.7 Flash
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...
Specifications
| Provider | StepFun |
| Model ID | stepfun/step-3.7-flash |
| Context window | 262K tokens (262,144) |
| Modalities | text + image + video → text |
| Tokenizer | Other |
| Router sources | OpenRouter, TrustedRouter, Vercel AI Gateway |
| Released | 2026-05-28 |
| Tracked since | 2026-11-14 |
| Popularity | Not in the ranked set |
| Reasoning | Mandatory |
Pricing (per million tokens)
| Variant | Source | API ID | Input | Output | Context |
|---|---|---|---|---|---|
| standard | OpenRouter | stepfun/step-3.7-flash | $0.2 | $1.15 | 262K |
| standard | TrustedRouter | stepfun-ai/step-3.7-flash | $0.211 | $1.21 | 256K |
| standard | Vercel AI Gateway | stepfun/step-3.7-flash | $0.2 | $1.15 | 256K |
Recorded price history
| Date | Input / 1M | Output / 1M | Context |
|---|---|---|---|
| 2026-11-14 | $0.2 | $1.15 | 256K |
| 2026-11-14 | $0.211 | $1.21 | 256K |
| 2026-11-14 | $0.211 | $1.21 | 262K |
| 2026-11-14 | $0.2 | $1.15 | 262K |
Capabilities
Vision (image input)Video inputAgentic tool useStructured outputsReasoning (always on)256K+ context windowControllable reasoningNo content storage (attested)Zero data retention
Trust flags come from TrustedRouter's per-provider attestations.
Supported parameters
frequency_penalty, include_reasoning, logprobs, max_tokens, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p