StepFun
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...
StepFun researchRank #134 across 566
Rank #113 across 222
Rank #104 across 371
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Higher bars mean stronger relative placement.
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 30.3 | #134 |
| Artificial Analysis Coding Index | coding | 39.6 | #113 |
| GPQA | reasoning | 80.9% | #136 |
| Humanity's Last Exam | reasoning | 19.9% | #105 |
| SciCode |
| coding, reasoning |
| 40.0% |
| #137 |
| Output Speed | speed | 385.2 tok/s | #5 |
| Time to First Token | speed | 0.67s | #78 |
| Blended Price | cost | $0.438/M | #104 |
| Input Price | cost | $0.200/M | #84 |
| Output Price | cost | $1.15/M | #121 |
| Value Index | cost, overall | 69.2 | #53 |