UTC

Step 3.7 Flash

StepFun (China) · open weights · served by 21 providers · tool calling, reasoning, image input, video input

stepfun/step-3.7-flash

Weights
Open
Released
2026-05-29
Knowledge cutoff
2026-03-01
Context window
256K
Hugging Face repository

stepfun-ai/Step-3.7-Flash at commit 5f6244077a, last changed 2026-06-03

Licence
apache-2.0
Access
Open download
Library
transformers
Formats
.safetensors
  • Loading it runs code from the repository (trust_remote_code). Read that code, and pin the commit.
  • No model signature file is published with the weights.

What the Hub's scanners report

26 weight files, 25 with a scan result. No scanner reports an issue with them.

The model at this commit, its 26 weight files with their SHA-256.

Facts from the Hugging Face Hub and its file scanners (Protect AI, Cisco, Hugging Face picklescan, VirusTotal, JFrog), as of 2026-11-18. ShieldMap does not copy model cards.

Who serves it
  • AIHubMixstep-3.7-flash
  • CommandCodestepfun/Step-3.7-Flash
  • Deep Infrastepfun-ai/Step-3.7-Flashdeprecated
  • Eden AIdeepinfra/stepfun-ai/Step-3.7-Flash
  • EmpirioLabs AIstep-3-7-flash
  • Hugging Facestepfun-ai/Step-3.7-Flash
  • Kenaristep-3-7-flash
  • Kilo Gatewaystepfun/step-3.7-flash
  • DevPass (LLM Gateway)step-3.7-flash
  • LLM Gatewaynovita/step-3.7-flash
  • NanoGPTstepfun/step-3.7-flash:thinking
  • Novita AIstepfun/step-3.7-flash
  • OpenRouterstepfun/step-3.7-flash
  • Requestystep-3.7-flash
  • StepFun (China)step-3.7-flash
  • StepFun (Global)step-3.7-flash
  • StepFun Step Plan (Global)step-3.7-flash
  • StepFun Step Plan (China)step-3.7-flash
  • UnoRouterstep-3.7-flash:freedeprecated
  • Vercel AI Gatewaystepfun/step-3.7-flash
  • ZenMuxstepfun/step-3.7-flash
Running it yourself

Open weights are loaded and served by code that has its own advisories. These inference servers and model loaders are on ShieldMap; load weights only from the publisher's own repository, and prefer safetensors over formats that can run code.

How supply-chain attacks on models work →

Sources: models.dev (MIT) and the LiteLLM model catalogue (MIT). Deprecation dates are as each platform published them and as LiteLLM records them; check the platform before you rely on one. Data as of 2026-11-22.