Model vs model
MiniMax M3 vs Step 3.7 Flash
Both are open-weights models that accept text and images. MiniMax M3 is the stronger pick: its Intelligence Index is 29.2 against 19.5, and it leads on every shared benchmark, most sharply on coding at 58.6 to 39.6. It costs more, though, at $0.52 per 1M tokens blended versus $0.44. Step 3.7 Flash suits high-volume work where price matters most.
At a glance
Longer bar means the bigger number.
Blended price · per 1M tokens, lower is better
Intelligence Index · Artificial Analysis, higher is better
Context window · tokens it can read at once
Where they sit among all 44 models
MiniMax M3 ranks 24th of the 43 models with an Intelligence Index, Step 3.7 Flash 32nd. Up and to the left is smarter for less money. Hover or tap a dot for its numbers, then click it to compare that model with MiniMax M3.
Which one to pick
Pick MiniMax M3 if
- you want the higher overall score (29.2 vs 19.5)
- you care about coding: 58.6 vs 39.6
- you care about Humanity's Last Exam (hard reasoning): 39 vs 21.4
- you need a longer context window (512K vs 256K tokens)
Pick Step 3.7 Flash if
- you want the lower bill ($0.44 vs $0.52 per 1M tokens blended)
How smart is each one?
Of the 8 scores both models have, MiniMax M3 is ahead on every one.
Every intelligence score we have
Out of 100, higher is better. MiniMax M3 Step 3.7 Flash
Overall
Intelligence Index
29.219.5
By area
Coding
58.639.6Agentic work
29.5Step 3.7 Flash: not scored yet
By field of work
Finance & accounting
30.1Step 3.7 Flash: not scored yet
Legal
31.5Step 3.7 Flash: not scored yet
Healthcare & medicine
27.8Step 3.7 Flash: not scored yet
Economics
43.7Step 3.7 Flash: not scored yet
Engineering
31.2Step 3.7 Flash: not scored yet
Strategy & operations
25Step 3.7 Flash: not scored yet
Benchmarks
Hard reasoning · Humanity's Last Exam
3921.4Science · GPQA Diamond
92.980.9Scientific coding · SciCode
47.143.9Terminal tasks · Terminal-Bench 4.0
2Step 3.7 Flash: not scored yet
Tool use · τ-Bench Banking
15.312Instruction following · IFBench
82.967.3Long documents · AA-LCR
8373.7
Our tests
Rule that has to last 4 turns · our test, % of tries passed
33Step 3.7 Flash: not scored yet
Three bullets, no letter e · our test, % of tries passed
100Step 3.7 Flash: not scored yet
The specs
| Spec | MiniMax M3 | Step 3.7 Flash |
|---|---|---|
| Input price | $0.30 per 1M | $0.20 per 1M |
| Output price | $1.20 per 1M | $1.15 per 1M |
| Cost per test task | $0.5076 | Not tested |
| Context window | 512K | 256K |
| Max output | 512K | 256K |
| Accepts | Text, Image, PDF | Text, Image |
| Open weights | Yes | Yes |
| Lab | MiniMax | StepFun |
| Released | Jun 2026 | May 2026 |
| Knowledge cutoff | Not published | Mar 2026 |
Blended price weighs 3 parts input to 1 part output. List prices from models.dev. Cost per test task from Artificial Analysis.
Questions
Is MiniMax M3 or Step 3.7 Flash cheaper?
Step 3.7 Flash is cheaper. It charges $0.20 per 1M input tokens and $1.15 per 1M output tokens, against $0.30 and $1.20 for MiniMax M3. Blended 3 parts input to 1 part output, that is $0.44 against $0.52 per 1M tokens.
Which is smarter, MiniMax M3 or Step 3.7 Flash?
MiniMax M3 scores higher on the Artificial Analysis Intelligence Index, 29.2 against 19.5 for Step 3.7 Flash. The biggest differences are on Coding (58.6 vs 39.6) and Humanity's Last Exam (39 vs 21.4).
Which has the longer context window?
MiniMax M3 accepts 512K tokens, against 256K for Step 3.7 Flash.
Can I compare MiniMax M3 and Step 3.7 Flash on my own prompt?
Yes. ModelCompare sends one prompt to both models at once and shows each answer side by side with what it cost. You can try it free without an account.
Related comparisons
Where the numbers come from
Intelligence Index and benchmark scores are from Artificial Analysis, at each model’s highest effort, except SimpleQA, which is from Epoch AI, and the Uncensored score, which is from the UGI Leaderboard. The hallucination test is ours: questions with one exact answer, fake evidence pushed against right answers and requests that tempt a model to name packages that don’t exist. Prices are list prices from models.dev. Example answers come from our own test runs. Updated October 25, 2026.