Model vs model
Step 3.7 Flash vs Trinity Large Thinking
Step 3.7 Flash leads most on long documents, scoring 73.7 on AA-LCR against 38 for Trinity Large Thinking. That gap makes it the stronger pick for long-document work and overall quality, with an Intelligence Index of 19.5 against 10.8, at $0.44 per 1M tokens blended versus $0.39. Trinity Large Thinking suits cheaper, high-volume text-only jobs where long documents matter less.
At a glance
Longer bar means the bigger number.
Blended price · per 1M tokens, lower is better
Intelligence Index · Artificial Analysis, higher is better
Where they sit among all 44 models
Step 3.7 Flash ranks 32nd of the 43 models with an Intelligence Index, Trinity Large Thinking 40th. 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 Step 3.7 Flash.
Which one to pick
Pick Step 3.7 Flash if
- you want the higher overall score (19.5 vs 10.8)
- you care about AA-LCR (long documents): 73.7 vs 38
- you care about coding: 39.6 vs 25.8
- you need to send images
Pick Trinity Large Thinking if
- you want the lower bill ($0.39 vs $0.44 per 1M tokens blended)
How smart is each one?
Of the 8 scores both models have, Step 3.7 Flash is ahead on every one.
Every intelligence score we have
Out of 100, higher is better. Step 3.7 Flash Trinity Large Thinking
Overall
Intelligence Index
19.510.8
By area
Coding
39.625.8Agentic work
1Step 3.7 Flash: not scored yet
By field of work
Finance & accounting
10.5Step 3.7 Flash: not scored yet
Legal
11.3Step 3.7 Flash: not scored yet
Economics
16.6Step 3.7 Flash: not scored yet
Engineering
16.2Step 3.7 Flash: not scored yet
Strategy & operations
6.7Step 3.7 Flash: not scored yet
Benchmarks
Hard reasoning · Humanity's Last Exam
21.415.8Science · GPQA Diamond
80.975.2Scientific coding · SciCode
43.940.6Terminal tasks · Terminal-Bench 4.0
0.5Step 3.7 Flash: not scored yet
Tool use · τ-Bench Banking
125.8Instruction following · IFBench
67.356.3Long documents · AA-LCR
73.738
The specs
| Spec | Step 3.7 Flash | Trinity Large Thinking |
|---|---|---|
| Input price | $0.20 per 1M | $0.25 per 1M |
| Output price | $1.15 per 1M | $0.80 per 1M |
| Cost per test task | Not tested | Not tested |
| Context window | 256K | 262K |
| Max output | 256K | 80K |
| Accepts | Text, Image | Text |
| Open weights | Yes | Yes |
| Lab | StepFun | Arcee AI |
| Released | May 2026 | Apr 2026 |
| Knowledge cutoff | Mar 2026 | Not published |
Blended price weighs 3 parts input to 1 part output. List prices from models.dev.
Questions
Is Step 3.7 Flash or Trinity Large Thinking cheaper?
Trinity Large Thinking is cheaper. It charges $0.25 per 1M input tokens and $0.80 per 1M output tokens, against $0.20 and $1.15 for Step 3.7 Flash. Blended 3 parts input to 1 part output, that is $0.39 against $0.44 per 1M tokens.
Which is smarter, Step 3.7 Flash or Trinity Large Thinking?
Step 3.7 Flash scores higher on the Artificial Analysis Intelligence Index, 19.5 against 10.8 for Trinity Large Thinking. The biggest differences are on AA-LCR (73.7 vs 38) and Coding (39.6 vs 25.8).
Which has the longer context window?
They are about the same: Step 3.7 Flash accepts 256K tokens and Trinity Large Thinking 262K.
Can I compare Step 3.7 Flash and Trinity Large Thinking 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.