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.
Added May 29, 2026
Model weightsContext Window
262.1K
Max Output
256K
Input Price (Auto)
$0.20/1M
Output Price (Auto)
$1.15/1M
Cache Read (Auto)
$0.040/1M
Capabilities
Benchmarks
Benchmarks
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
19.5
Coding Index
39.6
Agentic work
T²-Bench Telecom (legacy)
Legacy fallback · Conversational AI agents in dual-control scenarios
98.5%
Better than 98% of models compared
Harvey LAB-AA
Legal agentic work criterion pass rate
72.7%
Better than 21% of models compared
GDPval-AA v2
Economically valuable tasks (Elo)
845 Elo
Better than 42% of models compared
Document reasoning
AA-LCR v1.1
Long context reasoning with updated grading
73.7%
Better than 72% of models compared
Reasoning
HLE
Humanity's Last Exam
21.4%
Better than 71% of models compared
IFBench
Instruction-following benchmark
67.3%
Better than 80% of models compared
CritPt
Research-level physics reasoning
2.3%
Coding
Terminal-Bench Hard (legacy)
Legacy fallback · Agentic coding and terminal use
35.6%
Better than 81% of models compared
SciCode
Python programming for scientific computing
43.9%
Better than 38% of models compared
Knowledge
AA-Omniscience Accuracy
Proportion of correctly answered questions
25.8%
AA-Omniscience Hallucination Rate
Rate of incorrect answers among non-correct responses
85.0%
Legacy benchmarks
GPQA Diamond (legacy)
Graduate-level scientific reasoning
80.9%
Better than 68% of models compared
AA-LCR (unversioned / legacy)
Long context reasoning evaluation
73.7%
Better than 73% of models compared
GDPval-AA (unversioned / legacy)
Economically valuable tasks
17.2%
Last updated Sep 20, 2026, 12:01 AM
Artificial AnalysisProviders
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
Related text models
Compare Step 3.7 Flash Thinking with similar models from the same provider or model family.
Step 5 Preview
stepfun/step-5-previewStep 5 Preview is StepFun's 600B sparse MoE frontier model for production-scale agents, activating 27B parameters per token. It is built for software engineering, long-horizon tool use, research, professional knowledge work, and finance, with native text, image, and video understanding and a 1M-token context window. ⚠️ Note: This model routes through StepFun, so privacy and logging guarantees may be limited.
Step 3.5 Flash 2603
stepfun-ai/step-3.5-flash-2603Step 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 3.5 Flash
stepfun-ai/step-3.5-flashStepFun'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.
MiMo V2.6 Flash Uncensored Thinking
xiaomi/mimo-v2.6-flash-uncensored:thinkingMiMo V2.6 Flash Uncensored with maximum thinking enabled. Supports a 1M-token context window and tool calling.
MiMo V2.6 Flash Uncensored
xiaomi/mimo-v2.6-flash-uncensoredA text-only, lower-refusal MiMo V2.6 Flash finetune with a 1M-token context window, separate reasoning output, and tool calling.
MiMo V2.6 Flash
xiaomi/mimo-v2.6-flashMiMo V2.6 Flash is Xiaomi's native omnimodal 309B-parameter mixture-of-experts model, activating 15B parameters per token. It balances intelligence, efficiency, and cost for coding, general agents, visual tasks, and cybersecurity, with text, image, video, and audio understanding and a 1M-token context window.