Step 3.7 Flash

May 2026
Step 3.7 Flash
Compare in
on
—per task?
$0.20 / $1.15per 1M in / outcache $0.04
112 tok/sfirst token in 2.5 s?
256Kcontext

Rankingat high effort?

Strong at

S
Tool calling#4 of 242
B
Reasoning#121 of 336Science#111 of 312Long context#99 of 314Instructions#63 of 245Vision#49 of 125
C
Agents & work tasks#137 of 177Knowledge#168 of 320

Alternatives?

ModelScoreCostPer taskWhat changes
CheaperMiniMax M3.1 Flashthinking+52.8-40%$0.22No strict JSON+1BetterGPT-6 Lunalow+28.1-54%$0.005Function calling only with none · Chat Completions+3MOpenMotif 3+47.5+100%$0.50 / $2Memory (4-bit): 121 GB → 189 GB+1CheaperClaude Haiku 5.5max+47.9-54%$0.21Remove sampling parameters+5CheaperGLM 5.3 FlashmaxOpen+44.8-46%$0.28Memory (4-bit): 121 GB → 193 GB+5

Run it yourself

estimate
Fits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✗ 48 GB GPU✗ 80 GB GPU
LicenceApache-2.0free for commercial use

History

4 versions
Apr 26Release dateAA Omniscience Index3.5 · Feb3.5 · Apr3.7
Apr 26Release dateAA Omniscience Index3.5 · Feb3.5 · Apr3.7

Benchmarks

all categories
Chat & Knowledge
Humanity's Last Exam21.4#121
AA Omniscience Index-37.3#169
CritPt (physics)2.3#105
AA Intelligence Index 4.3.2≈19.5#125
AA Omniscience · Accuracy25.8#107
AA Omniscience · Hallucination Rate85#202
AA-LCR (long context) 1.173.7#99
GPQA Diamond80.9#112
Coding
SciCode43.9#80
Terminal-Bench Hard (version unstated)35.6#41
Terminal-Bench 2.139.3#86
Agents & Tool Use
GDPval-AA (work tasks) 2.117.8#137
τ³-Banking · AA12#83
ITBench SRE30.3#30
APEX-Agents14.8#21
EnterpriseOps-Gym-AA29.1#36
τ²-bench · AA98.5#4
Writing & Language
IFBench (instructions)67.3#63
Vision & Documents
MMMU-Pro (vision)75.3#49

FAQ

How much does Step 3.7 Flash cost?›

$0.20 per 1M input tokens and $1.15 per 1M output tokens; cached input $0.04 per 1M.

Is Step 3.7 Flash free to use?›

There is no free API; the cheapest paid price is $0.20 per 1M input tokens. The weights are open, so you can run it yourself.

How good is Step 3.7 Flash, and what is it best at?›

#121 of 336 in Chat & Knowledge, #80 of 142 in Coding, #137 of 177 in Agents & Tool Use, #63 of 245 in Writing & Language, #49 of 125 in Vision & Documents. Strongest at: Tool calling.

Is Step 3.7 Flash better than Step 3.5 Flash?›

On AA Omniscience Index, Step 3.7 Flash scores -37.3 vs -44.2 — higher.

How can I use Step 3.7 Flash?›

Through the API of DeepInfra, Featherless, Novita, StepFun.

What is the context window of Step 3.7 Flash?›

256K tokens, with up to 230K output tokens.

When was Step 3.7 Flash released? Is there a newer version?›

Released 2026-05-29. It is the current version in its line.

Is Step 3.7 Flash open source?›

Yes, the weights are open (Apache-2.0 licence).

Can I run Step 3.7 Flash locally?›

The weights are open and can be self-hosted; hardware needs are not confirmed yet.

Can Step 3.7 Flash read images?›

It accepts text and images, video as input.

What is Step 3.7 Flash?›

Step 3.7 Flash is an open-weights language model from StepFun, released May 2026, the successor to Step 3.5 Flash.

What are the best alternatives to Step 3.7 Flash?›

MiniMax M3.1 Flash (cheaper), GPT-6 Luna (better), Motif 3 (open weights).