Step 3.7 Flash
stepfun/step-3.7-flash-20260528
StepFun: Step 3.7 Flash is a model from stepfun that accepts text, image, and video inputs and generates text output. It supports tools and reasoning, provides open weights, and offers a context length of 262,144 tokens. The pricing is $0.20 per million input tokens and $1.15 per million output tokens.
- tool use
- structured outputs
- reasoning
- open-weight
At a glance
Specifications
What the API gives you
- Context window
- 262K tokens
- Max output
- 230K tokens
- Input price
- $0.20 / 1M tokens
- Output price
- $1.15 / 1M tokens
- Cache read
- $0.040 / 1M tokens
- Modalities
- text, image, video → text
- Knowledge cutoff
- —
- Released
- May 28, 2026
Measured evidence
Public benchmark record
- Design Arena Coding
- 1203 Elo2026-07-28
- Design Arena Coding
- 1203 Elo2026-07-27
- Design Arena Coding
- 1203 Elo2026-07-30
- Design Arena Coding
- 1203 Elo2026-07-29
- Design Arena Coding
- 1203 Elo2026-07-31
- Design Arena Coding
- 1203 Elo2026-08-02
- Design Arena Coding
- 1203 Elo2026-08-05
- Design Arena Coding
- 1203 Elo2026-08-01
- Design Arena Coding
- 1202 Elo2026-08-09
- Design Arena Coding
- 1203 Elo2026-08-03
- Design Arena Coding
- 1202 Elo2026-08-04
- Design Arena Coding
- 1201 Elo2026-08-14
- Design Arena Coding
- 1203 Elo2026-08-06
- Design Arena Coding
- 1202 Elo2026-08-10
- Design Arena Coding
- 1203 Elo2026-08-07
- Design Arena Coding
- 1196 Elo2026-08-18
- Design Arena Coding
- 1202 Elo2026-08-08
- Design Arena Coding
- 1195 Elo2026-08-15
- Design Arena Coding
- 1202 Elo2026-08-11
- Design Arena Coding
- 1202 Elo2026-08-12
- Design Arena Coding
- 1201 Elo2026-08-13
- Design Arena Coding
- 1194 Elo2026-08-19
- Design Arena Coding
- 1203 Elo2026-08-27
- Design Arena Coding
- 1194 Elo2026-08-20
- Design Arena Coding
- 1195 Elo2026-08-16
- Design Arena Coding
- 1202 Elo2026-08-21
- Design Arena Coding
- 1195 Elo2026-08-17
- Design Arena Coding
- 1202 Elo2026-08-22
- Design Arena Coding
- 1202 Elo2026-08-23
- Design Arena Coding
- 1202 Elo2026-08-28
- Design Arena Coding
- 1202 Elo2026-08-24
- Design Arena Coding
- 1203 Elo2026-08-25
- Design Arena Coding
- 1203 Elo2026-08-26
- Design Arena Coding
- 1203 Elo2026-08-29
- Design Arena Coding
- 1202 Elo2026-08-30
- Design Arena Coding
- 1202 Elo2026-08-31
- Design Arena Coding
- 1202 Elo2026-09-01
- Design Arena Coding
- 1201 Elo2026-09-02
- Design Arena Coding
- 1200 Elo2026-09-03
- Design Arena Coding
- 1199 Elo2026-09-04
- Design Arena UI
- 1203 Elo2026-07-28
- Design Arena UI
- 1202 Elo2026-07-27
- Design Arena UI
- 1201 Elo2026-07-30
- Design Arena UI
- 1202 Elo2026-07-29
- Design Arena UI
- 1201 Elo2026-07-31
- Design Arena UI
- 1202 Elo2026-08-02
- Design Arena UI
- 1202 Elo2026-08-01
- Design Arena UI
- 1204 Elo2026-08-05
- Design Arena UI
- 1204 Elo2026-08-09
- Design Arena UI
- 1202 Elo2026-08-03
- Design Arena UI
- 1204 Elo2026-08-04
- Design Arena UI
- 1202 Elo2026-08-14
- Design Arena UI
- 1204 Elo2026-08-06
- Design Arena UI
- 1205 Elo2026-08-07
- Design Arena UI
- 1204 Elo2026-08-10
- Design Arena UI
- 1203 Elo2026-08-08
- Design Arena UI
- 1202 Elo2026-08-18
- Design Arena UI
- 1206 Elo2026-08-11
- Design Arena UI
- 1200 Elo2026-08-15
- Design Arena UI
- 1203 Elo2026-08-12
- Design Arena UI
- 1203 Elo2026-08-13
- Design Arena UI
- 1202 Elo2026-08-19
- Design Arena UI
- 1201 Elo2026-08-16
- Design Arena UI
- 1202 Elo2026-08-20
- Design Arena UI
- 1201 Elo2026-08-27
- Design Arena UI
- 1202 Elo2026-08-17
- Design Arena UI
- 1199 Elo2026-08-21
- Design Arena UI
- 1199 Elo2026-08-22
- Design Arena UI
- 1200 Elo2026-08-23
- Design Arena UI
- 1200 Elo2026-08-24
- Design Arena UI
- 1201 Elo2026-08-28
- Design Arena UI
- 1201 Elo2026-08-25
- Design Arena UI
- 1201 Elo2026-08-26
- Design Arena UI
- 1201 Elo2026-08-29
- Design Arena UI
- 1200 Elo2026-08-30
- Design Arena UI
- 1200 Elo2026-08-31
- Design Arena UI
- 1200 Elo2026-09-01
- Design Arena UI
- 1199 Elo2026-09-02
- Design Arena UI
- 1199 Elo2026-09-03
- Design Arena UI
- 1200 Elo2026-09-04
- Design Arena Website
- 1212 Elo2026-07-28
- Design Arena Website
- 1212 Elo2026-07-30
- Design Arena Website
- 1212 Elo2026-07-27
- Design Arena Website
- 1213 Elo2026-07-29
- Design Arena Website
- 1212 Elo2026-07-31
- Design Arena Website
- 1212 Elo2026-08-02
- Design Arena Website
- 1211 Elo2026-08-01
- Design Arena Website
- 1212 Elo2026-08-05
- Design Arena Website
- 1212 Elo2026-08-03
- Design Arena Website
- 1211 Elo2026-08-09
- Design Arena Website
- 1211 Elo2026-08-04
- Design Arena Website
- 1210 Elo2026-08-14
- Design Arena Website
- 1212 Elo2026-08-06
- Design Arena Website
- 1211 Elo2026-08-07
- Design Arena Website
- 1210 Elo2026-08-10
- Design Arena Website
- 1211 Elo2026-08-08
- Design Arena Website
- 1202 Elo2026-08-18
- Design Arena Website
- 1211 Elo2026-08-11
- Design Arena Website
- 1202 Elo2026-08-15
- Design Arena Website
- 1211 Elo2026-08-12
- Design Arena Website
- 1210 Elo2026-08-13
- Design Arena Website
- 1202 Elo2026-08-19
- Design Arena Website
- 1201 Elo2026-08-16
- Design Arena Website
- 1202 Elo2026-08-20
- Design Arena Website
- 1213 Elo2026-08-27
- Design Arena Website
- 1202 Elo2026-08-17
- Design Arena Website
- 1211 Elo2026-08-21
- Design Arena Website
- 1211 Elo2026-08-22
- Design Arena Website
- 1212 Elo2026-08-23
- Design Arena Website
- 1212 Elo2026-08-24
- Design Arena Website
- 1213 Elo2026-08-28
- Design Arena Website
- 1213 Elo2026-08-25
- Design Arena Website
- 1213 Elo2026-08-26
- Design Arena Website
- 1212 Elo2026-08-29
- Design Arena Website
- 1212 Elo2026-08-30
- Design Arena Website
- 1212 Elo2026-08-31
- Design Arena Website
- 1212 Elo2026-09-01
- Design Arena Website
- 1211 Elo2026-09-02
- Design Arena Website
- 1210 Elo2026-09-03
- Design Arena Website
- 1209 Elo2026-09-04
LMArena data licensed CC-BY-4.0. Aider (Apache-2.0) and SWE-bench (MIT) are open. Design Arena uses blind human preference.
Open-weight model on Hugging Face.
Task evidence
How broad is the support?
Each status is exact to this model version and the selected task. Official release claims remain provisional.
- Best LLM for Frontend & UIProvisional evidence
60% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Coding or LMArena WebDev. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for CodingProvisional evidence
24% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 8 of 8 configured evaluation feeds currently have results.
- Best Cheap LLM for High VolumeProvisional evidence
17% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Text. 3 of 4 configured evaluation feeds currently have results.
- Cheapest Capable LLMProvisional evidence
17% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Text. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Fast SummariesProvisional evidence
14% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Text. 3 of 4 configured evaluation feeds currently have results.
- Best Value LLMProvisional evidence
14% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Text. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for High-Volume ExtractionProvisional evidence
13% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Data Analysis or Route reliability. 2 of 4 configured evaluation feeds currently have results.
- Best LLM for Code CompletionProvisional evidence
13% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for Literary TranslationProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for Everyday TranslationProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for TranslationProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for Debugging & Code ReviewProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 6 of 6 configured evaluation feeds currently have results.
- Best LLM for Document SummarizationProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from FACTS Grounding or LiveBench Instruction Following. 4 of 6 configured evaluation feeds currently have results.
- Best LLM for Tool & Function CallingProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for Business WritingProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Creative Writing. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Logic & Problem SolvingProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Marketing CopyProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Creative Writing. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for SummarizationProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from FACTS Grounding or LiveBench Instruction Following. 4 of 6 configured evaluation feeds currently have results.
- Best LLM for Chat & RoleplayProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Creative WritingProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Computer & Browser UseProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for OCR & DocumentsProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.
- Best LLM for Image UnderstandingProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Reasoning or LMArena Vision. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Vision & DocumentsProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.
- Best LLM for WritingProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Grounded RAGProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from CRAG or FACTS Grounding. 2 of 5 configured evaluation feeds currently have results.
- Best LLM for Roleplay & CharacterProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Charts & DiagramsProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.
- Best LLM for Document ParsingProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.
- Best LLM for a General AssistantProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.
- Best Open-Weight LLM for Self-HostingProvisional evidence
8% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Instruction Following or LMArena Text. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Massive ContextProvisional evidence
6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Long Query. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Hermes AgentProvisional evidence
6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for OpenHandsProvisional evidence
6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for GooseProvisional evidence
6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for RAG & Long ContextProvisional evidence
6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from FACTS Grounding or LMArena Document. 4 of 6 configured evaluation feeds currently have results.
- Best LLM for OpenClawProvisional evidence
6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for JSON & Schema OutputProvisional evidence
6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LiveBench Data Analysis. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for Structured Data ExtractionProvisional evidence
6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LiveBench Data Analysis. 4 of 5 configured evaluation feeds currently have results.
- Best Local LLM for CodingProvisional evidence
6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Competition MathProvisional evidence
5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Mathematics or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Agentic CodingProvisional evidence
5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Design Arena Full-stack or Design Arena Web-apps. 8 of 8 configured evaluation feeds currently have results.
- Best LLM for Autonomous AgentsProvisional evidence
5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Frontier-Bench or LiveBench Agentic Coding. 7 of 7 configured evaluation feeds currently have results.
- Best LLM for Math & ReasoningProvisional evidence
5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Mathematics or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for AI AgentsProvisional evidence
5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LiveBench Agentic Coding. 7 of 7 configured evaluation feeds currently have results.
Recommendation coverage
Where this model appears
Not currently included in a published recommendation. The task evidence panel above says exactly what is confirmed and what is still missing.
Catalog data via OpenRouter and models.dev. Last checked within the past 6 hours.