# Step 3.5 Flash vs Trinity Large Thinking

> Step 3.5 Flash is the stronger model overall, scoring 42.3 to 38.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/step-3-5-flash-vs-trinity-large-thinking
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. Step 3.5 Flash scores higher in 7 categories and Trinity Large Thinking in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Step 3.5 Flash leads 42.4 to 34.1.
- The biggest single-benchmark swing is NYT Connections (extended): 28.4% for Step 3.5 Flash and 16.5% for Trinity Large Thinking.
- Step 3.5 Flash is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.25 / $0.80 for Trinity Large Thinking.
- Trinity Large Thinking accepts more context: 262K tokens versus 256K.

## Snapshot

| | Step 3.5 Flash | Trinity Large Thinking |
|---|---|---|
| Provider | StepFun | Arcee AI |
| Noometry Index | 42.3 | 38.6 |
| Rank | 116 | 185 |
| Context | 256K | 262K |
| Input $/M | $0.10 | $0.25 |
| Output $/M | $0.30 | $0.80 |
| Weights | Open | Open |

## Coding

- Step 3.5 Flash: 42.4 (#105)
- Trinity Large Thinking: 34.1 (#244)

| Benchmark | Step 3.5 Flash | Trinity Large Thinking |
|---|---|---|
| LMArena Coding | 1436 | 1381 |
| LMArena WebDev | — | 1238 |
| SciCode | — | 36.1% |

## Reasoning

- Step 3.5 Flash: 22.2 (#202)
- Trinity Large Thinking: 16.9 (#298)

| Benchmark | Step 3.5 Flash | Trinity Large Thinking |
|---|---|---|
| NYT Connections (extended) | 28.4% | 16.5% |
| LMArena Hard Prompts | 1411 | 1350 |
| CritPt | — | 0.9% |
| Thematic Generalization | — | 41.6% |
| Surface Evolver Bench | — | 15.6% |

## Math

- Step 3.5 Flash: 42.6 (#84)
- Trinity Large Thinking: 37.6 (#149)

| Benchmark | Step 3.5 Flash | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1408 | 1366 |
| MathArena Final-Answer Competitions | 66.8% | — |

## Knowledge

- Step 3.5 Flash: 39.6 (#132)
- Trinity Large Thinking: 40.9 (#113)

| Benchmark | Step 3.5 Flash | Trinity Large Thinking |
|---|---|---|
| LMArena Expert | 1421 | 1360 |
| Vectara Hallucination Rate | — | 6.9% |

## Multilingual

- Step 3.5 Flash: 50.5 (#119)
- Trinity Large Thinking: 46.2 (#160)

| Benchmark | Step 3.5 Flash | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1385 | 1325 |
| LMArena Chinese | 1447 | 1373 |
| LMArena French | 1421 | 1374 |
| LMArena German | 1405 | 1356 |
| LMArena Japanese | 1354 | 1311 |
| LMArena Korean | 1352 | 1306 |
| LMArena Russian | 1385 | 1337 |
| LMArena Spanish | 1419 | 1357 |

## Instruction Following

- Step 3.5 Flash: 73.1 (#124)
- Trinity Large Thinking: 70.5 (#162)

| Benchmark | Step 3.5 Flash | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1385 | 1334 |

## Long Context

- Step 3.5 Flash: 42.8 (#117)
- Trinity Large Thinking: 41.3 (#144)

| Benchmark | Step 3.5 Flash | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1402 | 1355 |

## Writing & Preference

- Step 3.5 Flash: 58.8 (#113)
- Trinity Large Thinking: 53.8 (#158)

| Benchmark | Step 3.5 Flash | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1403 | 1340 |
| LMArena Creative Writing | 1357 | 1320 |
| LMArena Multi-Turn | 1405 | 1342 |

## FAQ

### Is Step 3.5 Flash better than Trinity Large Thinking?

Step 3.5 Flash is the stronger model overall, scoring 42.3 to 38.6 on the Noometry Index.

### Which is cheaper, Step 3.5 Flash or Trinity Large Thinking?

Step 3.5 Flash is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Trinity Large Thinking lists at $0.25 and $0.80.

### Is Step 3.5 Flash or Trinity Large Thinking better for coding?

Step 3.5 Flash scores higher on coding benchmarks: 42.4 versus 34.1 in the Noometry coding category.

### Which has the bigger context window?

Trinity Large Thinking does, with 262K tokens against 256K.

### How many benchmarks do Step 3.5 Flash and Trinity Large Thinking share?

18 benchmarks have published results for both models. Step 3.5 Flash has 19 scored results on Noometry and Trinity Large Thinking has 24.