Nano GPT logo
NanoGPT
Provider logo

Step 3.5 Flash

stepfun-ai/step-3.5-flash
BackTry Model
Provider logo

Step 3.5 Flash

stepfun-ai/step-3.5-flash
BackTry Model

StepFun's most capable open-source reasoning model with visible reasoning traces. Built on a sparse Mixture-of-Experts architecture with 196B total parameters and only 11B active per token, it achieves frontier-level performance in math, logic, and agentic coding while reaching up to 350 tokens/sec. Supports 256K context. NOTE: This model runs via StepFun, which may log and train on your prompts.

Added Feb 2, 2026

Model weights

Context Window

256.0K

Max Output

256.0K

Avg output tokens (7d)

2.2K tokens

80%

Input Price (Auto)

$0.10/1M

Output Price (Auto)

$0.30/1M

Cache Read (Auto)

$0.050/1M

Capabilities

Benchmarks

Performance metrics and benchmarks

Sourced from Artificial Analysis.

Intelligence Index

26.5

Better than 65% of models compared

Reasoning

GPQA Diamond

Graduate-level scientific reasoning

83.1%

Better than 76% of models compared

HLE

Humanity's Last Exam

21.1%

Better than 76% of models compared

IFBench

Instruction-following benchmark

64.6%

Better than 76% of models compared

T²-Bench Telecom

Conversational AI agents in dual-control scenarios

94.4%

Better than 92% of models compared

AA-LCR

Long context reasoning evaluation

48.0%

Better than 52% of models compared

CritPt

Research-level physics reasoning

2.5%

Coding

SciCode

Python programming for scientific computing

40.4%

Better than 72% of models compared

Terminal-Bench Hard

Agentic coding and terminal use

27.3%

Better than 70% of models compared

Knowledge

AA-Omniscience Accuracy

Proportion of correctly answered questions

23.6%

AA-Omniscience Hallucination Rate

Rate of incorrect answers among non-correct responses

85.7%

Last updated Aug 7, 2026, 12:00 AM

Artificial Analysis

Providers

Auto routing is available for this model. Explicit provider selection is not available.

Loading provider options…

Compare Step 3.5 Flash with similar models from the same provider or model family.

Step 3.5 Flash 2603

stepfun-ai/step-3.5-flash-2603

Step 3.5 Flash 2603 is optimized for high-frequency agentic and coding workflows with improved token efficiency and faster reasoning. NOTE: This model runs via StepFun, which may log and train on your prompts.

Step 3.7 Flash Thinking

stepfun/step-3.7-flash:thinking

Step 3.7 Flash Thinking is StepFun's high-efficiency multimodal MoE model with visible reasoning enabled for deeper agentic coding, long-context reasoning, tool use, and native image/video understanding. ⚠️ Note: This model routes through StepFun, so privacy and logging guarantees may be limited.

Step R1 V Mini

step-r1-v-mini

Step-R1-V-Mini, which supports image and text input, text output, has good instruction following and general capabilities, can perceive images with high precision and complete complex reasoning tasks.

DeepSeek V4 Flash Vision Exp

deepseek/deepseek-v4-flash-vision-exp

An experimental vision-enabled DeepSeek V4 Flash model that adds image understanding while retaining the text, reasoning, coding, tool-calling, and agent capabilities of the base model. This route is served directly by DeepSeek, so privacy and logging guarantees are limited.

Gemini 3.7 Flash

google/gemini-3.7-flash

Google's frontier-performance Flash model for multimodal and agentic workloads, including coding, tool use, image understanding, PDF and document extraction, audio, and video. Google reports 65.3% on DeepSWE v1.1, up from 49.0% for Gemini 3.6 Flash, and 34% on GDP.pdf, up from 14%.

DeepSeek V4 Flash Latest

deepseek/deepseek-v4-flash-latest

Compatibility alias that routes to the newest dated DeepSeek V4 Flash release.