← Model list
Step 3.7 Flash
stepfun·stepfun/step-3.7-flash·GA·Open weights·NEW
Newer StepFun flash model for faster agents, coding, and multimodal prompts
At a glance
Official price$0.185 / $1.11 per 1M
Cache read $0.037 · StepFun (Global)
Lowest paid$0.18 / $1.03
Requesty Gateway · 1.1× spread
2 more $0 channels
Context256,000
⚠ Providers report 256,000–262,144; the table below is authoritative
Output limit256,000
Capabilities
✓ Reasoning✓ Tool use✓ Structured output✓ Temperature✓ Attachments
Modalities
TextImageVideo
Knowledge cutoff2026-03-01
Released / updated2026-05-29 / 2026-05-29
Quality & performanceArtificial Analysis · Intelligence Index v4.1
Intelligence30.9
Value78
Coding39.6
Agentic21.7
Output speed117 tok/s
TTFT2.74 s
Cost per task$0.0913
Value formulaIQ 30.9 ÷ min blended $0.394 = 78
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
Available at 16 providers12 with public prices · 2 free · 2 subscription-covered
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| Nvidia | First-party | Free | — | — | 256,000 | 16,384 | ||
| UnoRouter step-3.7-flash:free | Gateway | Free | — | — | 256,000 | 256,000 | ||
| Requesty | Gateway | $0.18 | $1.03 | $0.036 | — | 262,144 ⚠ | 256,000 | |
| StepFun (Global)Official | First-party | $0.185 | $1.11 | $0.037 | — | 256,000 | 256,000 | |
| StepFun (China)Official | First-party | $0.185 | $1.11 | $0.037 | — | 256,000 | 256,000 | |
| Ambient | Gateway | $0.19 | $1.14 | $0.03 | $0 | 262,144 ⚠ | 262,144 | |
| Deep Infra stepfun-ai/Step-3.7-Flash | Cloud | $0.20 | $1.15 | $0.04 | — | 262,144 ⚠ | 256,000 | |
| Hugging Face stepfun-ai/Step-3.7-Flash | Cloud | $0.20 | $1.15 | — | — | 262,144 ⚠ | 256,000 | |
| OpenRouter | Gateway | $0.20 | $1.15 | $0.04 | — | 262,144 ⚠ | 256,000 | |
| ZenMux | Gateway | $0.20 | $1.15 | — | — | 256,000 | 256,000 | |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Your usage cost
1Requesty$70.47
2StepFun (Global) · Official$74.74
3StepFun (China) · Official$74.74
4Ambient$75.80
5Deep Infra$78.30
6OpenRouter$78.30
2 more channels offer $0 (Nvidia, UnoRouter); free tiers usually have rate limits and no SLA, excluded from ranking.
Switch to Requesty to save $4.27/mo (6%). The gap is small, so staying on the official channel is fine.
Note: this is a gateway; verify availability and rate limits yourself.
Benchmark11 items
| Name | Conditions | Score | Metric | Source |
|---|---|---|---|---|
| SWE-Bench Pro | — | 56.3 | resolve rate | Source ↗ |
| SWE-Bench Verified | — | 76.5 | resolved | Source ↗ |
| Terminal-Bench | v2.1 | 59.6 | success rate | Source ↗ |
| Humanity's Last Exam | variant: with tools | 47.2 | accuracy | Source ↗ |
| BrowseComp | — | 75.8 | accuracy | Source ↗ |
| Toolathlon | — | 49.5 | success rate | Source ↗ |
| GDPval | — | 45.8 | wins or ties | Source ↗ |
| ClawEval | v1.1 | 67.1 | pass^3 | Source ↗ |
| Artificial Analysis Coding Index | — | 37.1 | index | Source ↗ |
| SciCode | — | 40 | percent correct | Source ↗ |
| Terminal-Bench Hard | — | 35.6 | success rate | Source ↗ |
The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.
Reasoning control
effort = minimaleffort = loweffort = mediumeffort = higheffort = xhigheffort = maxeffort = nonebudget_tokens
2 / 16 providers expose no reasoning control (reasoning_options: []).
Related models
DeepSeek V4 Flashcheaper alternative$0.14 / $0.28GPT-5 Nanocheaper alternative$0.05 / $0.40GLM-4.7-Flashcheaper alternative$0 / $0MiMo-V2.5cheaper alternative$0.14 / $0.28
Price historyone sample accumulated per data sync
Input listOutput listMin blended