Models/Step 3.7 Flash

Text / stepfun-ai

Step 3.7 Flash

Step 3.7 Flash is a sparse Mixture-of-Experts (MoE) vision-language model with 198B total parameters, combining a 196B-parameter language backbone and a 1.8B-parameter vision encoder. It is licensed under Apache 2.0, uses the image-text-to-text pipeline, and is associated with the safetensors format and Transformers library. The model activates approximately 11B parameters per token and is described as supporting a 256k context window with three selectable reasoning levels (low, medium, high). It is intended for high-frequency production workloads and agentic workflows involving perception, search, and reasoning.

EstimatedRepository opened May 23, 2026Source checked 9/24/2026Version: 5f624407

LOCALRENTED GPUOPEN WEIGHTS

Freshness 52 out of 100. Repository age: 170 days old, May 23, 2026. Measures age, not quality.

Popularity 57 out of 100. 10,332 downloads in 30 days; 463 lifetime likes. Checked Nov 8, 2026. Measures adoption, not quality.

What do these scores mean?

Two ways to compare models. Neither score measures answer quality, benchmark performance, or whether a model fits your machine.

Freshness · release age

A new release starts at 100. The score halves every 180 days: 100 × 2^(−age in days / 180). When the release date is unavailable, a repository date is a fallback—not proof of when the model was released. Missing dates stay unscored.

Check the date source

Popularity · adoption

Downloads over the past 30 days carry 80% of the score; lifetime likes carry 20%. Both use a logarithmic scale, capped at 10 million downloads and 100,000 likes. If likes are unavailable, downloads carry all the weight. Missing download statistics stay unscored.

round(80 × min(1, log10(1 + downloads) / 7) + 20 × min(1, log10(1 + likes) / 5))

Check the statistics source

Scores as of Nov 9, 2026. Statistics checked Nov 8, 2026. Statistics are marked for refresh after 72 hours.Full scoring method

At a glance

Parameters: 201.37B
Architecture: step3p7
License: apache-2.0
Disk space: 375.1 GB
Software: SGLang

View the model source

MY HARDWARE

Will it run on your machine?

Save your machine to see a personalized rating and its reasoning.

Add my hardware

Best for

Parsing large financial reports or other dense documents in a single pass.
Running multi-step search loops with cross-source verification.
Operating concurrent coding agents in high-throughput pipelines.
Processing dense visual interfaces such as UI wireframes, application GUIs, and data charts into structured code.
Interacting with external APIs and executing long-horizon autonomous workflows.

Tradeoffs

Some benchmark areas, such as Terminal-Bench 2.1 and GDPVal-AA, show clear opportunities for optimization compared to top performers.
The model requires substantial memory and compute due to its 198B total parameters, though only about 11B are activated per token.
Deployment on local or workstation environments is mentioned but detailed hardware requirements are not provided.
The model card primarily covers English language support; broader language capabilities are not specified.
Pricing and availability are subject to platform-specific terms and may change; self-hosting details are not fully provided.

Ways to run it

SGLang · Linux

Repository-specific command in publisher documentation; confirm dependencies and hardware in the source.

sglang serve --model-path stepfun-ai/Step-3.7-Flash  --tp 8  --reasoning-parser step3p5  --tool-call-parser step3p5  --enable-multimodal  --speculative-algorithm EAGLE  --speculative-num-steps 3  --speculative-eagle-topk 1  --speculative-num-draft-tokens 4  --enable-multi-layer-eagle  --trust-remote-code  --host 0.0.0.0 \

Official instructions

Explore its uses

SAFETENSORSIMAGE-TEXT-TO-TEXT

Keep exploring

Text

Prism ML

Bonsai 2 · 27B

A compressed 27B-class reasoning model with publisher-provided GGUF packs and a dedicated llama.cpp fork for CUDA, Metal, and CPU.

Freshness 82 out of 100. Release age: 53 days old, Sep 17, 2026. Measures age, not quality.

MODEL SIZE27.36B

LOCALRENTED GPU

Released Sep 17, 2026

Text

DeepSeek

DeepSeek V4.1 Flash

An open-weight model for reasoning, coding, agents, and image understanding. DeepSeek V4.1 Flash combines a 552B backbone with sparse activation and up to one million tokens of context. Choose the original checkpoint or a community uncensored variant below.

Freshness 79 out of 100. Repository age: 60 days old, Sep 10, 2026. Measures age, not quality.

MODEL SIZE552B backbone · 8B active prefill / 16B active decode

LOCALRENTED GPU

Repository opened Sep 10, 2026

Text

Qwen

Qwen3.8

Explore 4 versions of Qwen3.8. Compare model sizes, packages, hardware requirements, and available setups.

4 versionsCompare sizes & formats

LOCALRENTED GPU

Compare versions →