Step 3.7 Flash
Step 3.7 Flash is worth evaluating for coding, rag, and agents when its provider route and context window match the workload.
Use it for
- Teams evaluating coding, rag, and agents
- Workloads that can use a 256k context window
- Buyers comparing 3 tracked provider routes
Do not use it for
- Workloads where another current model has stronger sourced task evidence
- Family
- Step
- Released
- 2026-05-29
- Context
- 256k
- Parameters
- 198B (11B active)
- Architecture
- Mixture of Experts
- Specialization
- general
- Openness
- Proprietary
- License
- ProprietaryCommercial use: conditional
- Weights
- Not released
- Code
- Unknown
- Training
- Pretrained
Cheapest of 3 routes · OpenRouter
About
Step 3.7 Flash is StepFun's open-weights multimodal Mixture-of-Experts model for agentic coding, tool use, long-context reasoning, image understanding, and video understanding. It combines a 196B-parameter language backbone with a 1.8B-parameter vision encoder, activates about 11B parameters per token, supports a 256K-token context window, and exposes low, medium, and high reasoning levels for speed/depth tradeoffs. StepFun reports leading open-model results on ClawEval-1.1, SimpleVQA with Search, and SWE-bench Pro at launch. Weights are available on Hugging Face under Apache 2.0.
Step 3.7 Flash is a proprietary model in the Step family. The structured metadata tracks a 256k-token context window, multimodal input, reasoning, function calling, tool use, and structured outputs. This page tracks provider routes through StepFun, OpenRouter, and NVIDIA NIM, with the cheapest tracked route listed at $0.2 input and $1.15 output per 1M tokens. Headline tracked benchmarks include ClawEval-1.1 67.1, SimpleVQA with Search Tool 79.2, and V* with Python 95.3.
Top use-case fit: coding, agents, and build tasks
Coding
Q/$ C1 relevant benchmark in the decision map.
RAG
Included by capability and metadata signals in the decision map.
Agents
Included by capability and metadata signals in the decision map.
Provider price ladder
Compare all 3Compare API pricing across 3 providers for input and output tokens, batch, and cached reads when available.
| Provider | Input / 1M | Output / 1M | Cache | Route |
|---|---|---|---|---|
| OpenRouter | $0.200 | $1.15 | - | Serverless |
| StepFun | $0.200 | $1.15 | read $0.040 | Serverless |
| NVIDIA NIM | - | - | - | ProvisionedPartial |
Available via routers & gateways(1)
Capabilities
Benchmark peer barsfor Coding
Benchmark scores(14)
| Benchmark | Score | Version | Source |
|---|---|---|---|
| ClawEval-1.1 | 67.1 | 1st among open models at release | https://static.stepfun.com/blog/step-3.7-flash/ |
| SimpleVQA with Search Tool | 79.2 | 1st at release (GPT-5.5: 79.1) | https://static.stepfun.com/blog/step-3.7-flash/ |
| V* with Python | 95.3 | 2nd at release (Kimi K2.6: 96.9) | https://huggingface.co/stepfun-ai/Step-3.7-Flash |
| SWE-bench Pro | 56.3 | 2nd at release (Claude Opus 4.7: 64.3, GPT-5.5: 58.6) | https://static.stepfun.com/blog/step-3.7-flash/ |
| Terminal-Bench 2.1 | 59.5 | Comparison: Step 3.5 Flash 53.37%, DeepSeek V4 Flash 62.0%, Gemini 3.5 Flash 76.2%, GPT-5.5 82.7%, Claude Opus 4.7 69.4% | https://static.stepfun.com/blog/step-3.7-flash/ |
| Toolathlon | 49.5 | — | https://huggingface.co/stepfun-ai/Step-3.7-Flash |
| Humanity's Last Exam | 47.2 | — | https://huggingface.co/stepfun-ai/Step-3.7-Flash |
| GDPval-AA | 45.8 | — | https://huggingface.co/stepfun-ai/Step-3.7-Flash |
| WorldVQA | 58.1 | Comparison: Kimi K2.6 55.98% | https://static.stepfun.com/blog/step-3.7-flash/ |
| HR-Bench 4K | 89.1 | Comparison: Kimi K2.6 91.25% | https://static.stepfun.com/blog/step-3.7-flash/ |
| Android Daily | 61.9 | Comparison: Gemini 3 Flash 63.21% | https://static.stepfun.com/blog/step-3.7-flash/ |
| DeepSearchQA | 92.8 | — | https://static.stepfun.com/blog/step-3.7-flash/ |
| BrowseComp | 75.8 | From official StepFun blog post (accuracy%) | https://static.stepfun.com/blog/step-3.7-flash/ |
| ResearchRubrics | 71.7 | — | https://static.stepfun.com/blog/step-3.7-flash/ |
Migration checks
No linked migration route is available for this model yet.
API versions
step-3.7-flashCompare Step 3.7 Flash with other models
Comparison and alternatives
Browse all comparisons →Frequently asked questions
What is the context window of Step 3.7 Flash?
Step 3.7 Flash has a context window of 256k tokens.
How much does Step 3.7 Flash cost?
Step 3.7 Flash pricing ranges from $0.20/1M to $0.2/1M input tokens depending on the provider.
When was Step 3.7 Flash released?
Step 3.7 Flash was released on 2026-05-29.
Which providers offer Step 3.7 Flash?
Step 3.7 Flash is available from 3 providers: StepFun, OpenRouter, NVIDIA NIM.
What benchmarks has Step 3.7 Flash been tested on?
Step 3.7 Flash has been evaluated on 14 benchmarks, including ClawEval-1.1, SimpleVQA with Search Tool, V* with Python, SWE-bench Pro, Terminal-Bench 2.1.
Cheapest of 3 routes · OpenRouter