Stepfun
Step 3.7 Flash
Released 2026-05-28
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...
You send / 1M
$0.200
tokens in
It replies / 1M
$1.15
tokens out
Memory
262K
tokens it can read at once
Typical request
$0.0008
1,000 words in, 400 out
Data privacy
No training or retention: Yes. At least one provider serves it with zero data retention.
| Provider | You send / 1M | It replies / 1M | Uptime | Privacy policy |
|---|---|---|---|---|
| Novita · fp8 | $0.200 | $1.15 | 100% | Read |
Based on OpenRouter and Vercel AI Gateway data. Confirm terms with the provider before sending sensitive data.
What it can do
- Reads: text, image, video
- Writes: text
- Reasoning: Yes
- Tool calling: Yes
- Longest reply: 230K tokens
- Repeated text is billed at $0.040 per 1M (prompt caching)
- Intelligence score: 19.5 / 100 (Artificial Analysis)
Similar models
- Qwen3.6 FlashQwen · $0.188 in / $1.13 out per 1M
- DeepSeek V4.1 Flash 1017DeepSeek · $0.300 in / $1.20 out per 1M
- Muse Glimmer 30BMeta · $0.300 in / $1.20 out per 1M
- Inkling SmallThinkingmachines · $0.450 in / $1.20 out per 1M
- GPT-5.6 Luna ProOpenAI · $0.200 in / $1.20 out per 1M
- GPT-5.6 LunaOpenAI · $0.200 in / $1.20 out per 1M
Appears in
Top No-Training AI ModelsTop No-Training Zero Data Retention AI ModelsTop No-Training Zero Data Retention Vision AI ModelsTop No-Training Zero Data Retention Reasoning AI ModelsTop No-Training Zero Data Retention Tool-Calling AI ModelsTop No-Training Vision AI ModelsTop No-Training Vision Reasoning AI ModelsTop No-Training Vision Tool-Calling AI Models