StepFun: Step 3.7 Flash
stepfun/step-3.7-flash
87th smartest of 182ReasoningToolsJSONVision
Intelligence
30.3
87th of 182
Design Elo
1211
Website
Speed
—
tokens/sec
Latency
—
first token
Input price
$0.200
96th cheapest
Context
262K
256K max out
How it compares
Smarter than52%
of all ranked models
Cheaper than71%
of all ranked models
Overview
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...
Benchmarks
independent · Artificial Analysis & Design ArenaArtificial Analysis76th percentile
Intelligence Index
30.3
Coding Index
39.6
GPQA Diamond
81%
Humanity's Last Exam
20%
SciCode
40%
Tau²-Bench (agentic)
99%
Design Arena · Elo9,928 tournaments
Website
1211
Data Viz
1205
3D
1178
svg
1116
Providers & pricing (3)
| Provider | In $/M | Out $/M | Context | Uptime |
|---|---|---|---|---|
| StepFunfp8 | $0.200 | $1.15 | 256K | 99.9% |
| DeepInfra | $0.200 | $1.15 | 262K | 99.3% |
| Novitafp8 | $0.200 | $1.15 | 262K | 100% |
Specifications
Context window262K
Max output256K
Knowledge cutoff—
Input modalitiestext, image, video
Output modalitiestext
Prompt caching—
Cache read price$0.040/M
ModeratedNo
Open weightsstepfun-ai/Step-3.7-Flash ↗
Step 3.7 Flash FAQ
How much does Step 3.7 Flash cost?
Step 3.7 Flash costs $0.200 per million input tokens and $1.15 per million output tokens via OpenRouter, making it 96th cheapest of 332 paid models.
How smart is Step 3.7 Flash?
Step 3.7 Flash scores 30.3 on the Artificial Analysis Intelligence Index, ranking 87th of 182 benchmarked models, with a GPQA Diamond score of 81%.
What is Step 3.7 Flash's context window?
Step 3.7 Flash supports a 262K-token context window and can output up to 256K tokens. It accepts text, image, video input.