Step 3.7 Flash

Name: Step 3.7 Flash
Author: StepFun

StepFunOpen WeightsPending Human Review

Step 3.7 Flash is an open-weight vision-language model for high-throughput agent execution. Its 198B total parameters combine a 196B sparse language backbone with a 1.8B vision encoder, activating about 11B parameters per token. It accepts images and text, supports a 256K context, and offers low, medium, and high reasoning levels. Training targets visual search, browser and terminal tool use, coding, and professional document workflows. API, web chat, and downloadable weights are available. Advisor-mode results depend on an additional larger model and should not be interpreted as standalone checkpoint performance.

2026-05-29

198B total, approximately 11B active

Mixture-of-Experts Transformer with vision encoder

Apache-2.0

Compare with other models

Specifications

Parameters: 198B total, approximately 11B active
Architecture: Mixture-of-Experts Transformer with vision encoder
License: Apache-2.0
Context Window: 262,144 tokens
Type: multimodal
Modalities: textimage

Benchmark Scores

Advanced Specifications

Model Family: Step
API Access: Available
Chat Interface: Available
Variants: Step-3.7-Flash

Capabilities & Limitations

Capabilities: visionreasoningagentic codingtool useweb search
Known Limitations: Full weights require high-memory deploymentAdvisor mode depends on a separate modelTool and search results depend on external infrastructure
Notable Use Cases: coding agentsfinancial document analysisvisual search workflows
Function Calling Support: Yes
Tool Use Support: Yes

Resources

Related Models

Step 5 Preview

StepFun

Step 5 Preview is StepFun’s hosted flagship for agentic coding, research, professional knowledge work, and financial analysis. Its sparse mixture-of-experts architecture has 600B total parameters and 27B active per token. The documented API accepts text, images, and video, generates text, and supports a 1M-token context with up to 64K output. Tool calls, JSON Schema output, caching, and low, medium, or high reasoning effort support sustained workflows in an external application. It was publicly announced on September 20, 2026; September is used because earlier reported API availability dates are not resolved by a first-party dated changelog. Announced October 15 weights are future availability at this audit cutoff.

Typemultimodal

Parameters600B total, 27B active

2026-09

Proprietary

Details Compare

Step 3.5 Flash

StepFun

Step 3.5 Flash is StepFun’s sparse language model optimized for reasoning, coding, and tool-driven agents. A 3:1 sliding-window/full-attention pattern reduces long-context inference cost while multi-token prediction increases throughput. The model activates 11B of 196B parameters and supports 256K context. Public access predates the February 12 technical-blog update; repositories and contemporaneous API catalogs establish the early-February release. The full weight footprint, context length, and selected reasoning mode affect deployment cost and latency.

Typetext

Parameters196B total, 11B active

2026-02-02

Open Weights

Details Compare

Step3 VL 10B

StepFun

Step3 VL 10B is StepFun’s compact vision-language reasoning model, combining an approximately 1.8B visual encoder with a language backbone. It targets visual STEM reasoning and language generation from images. PaCoRe mode combines evidence from sixteen parallel rollouts, so its scores should be separated from ordinary single-model sampling. Apache-licensed base and post-trained checkpoints support research and local evaluation. The full weight footprint, context length, and selected reasoning mode affect deployment cost and latency.