# GPT-4 Turbo vs Step 3.5 Flash

> Step 3.5 Flash is the stronger model overall, scoring 41.7 to 30.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-turbo-vs-step-3-5-flash
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. GPT-4 Turbo scores higher in 0 categories and Step 3.5 Flash in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 3.5 Flash leads 40.6 to 8.6.
- Step 3.5 Flash is cheaper at $0.10 / $0.30 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
- Step 3.5 Flash accepts more context: 256K tokens versus 128K.
- Step 3.5 Flash has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4 Turbo | Step 3.5 Flash |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 30.3 | 41.7 |
| Rank | 281 | 116 |
| Context | 128K | 256K |
| Input $/M | $10 | $0.10 |
| Output $/M | $30 | $0.30 |
| Weights | Proprietary | Open |

## Coding

- GPT-4 Turbo: 33.4 (#247)
- Step 3.5 Flash: 41.7 (#106)

| Benchmark | GPT-4 Turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1268 | 1436 |
| WeirdML | 18% | — |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 58.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |

## Agentic & Tool Use

- GPT-4 Turbo: —
- Step 3.5 Flash: —

| Benchmark | GPT-4 Turbo | Step 3.5 Flash |
|---|---|---|
| METR Time Horizons | 36.7% | — |

## Reasoning

- GPT-4 Turbo: 14.9 (#311)
- Step 3.5 Flash: 21.2 (#202)

| Benchmark | GPT-4 Turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1411 |
| SimpleBench | 25.1% | — |
| NYT Connections (extended) | — | 28.4% |
| Chess Puzzles | 6% | — |
| DTBench | 61.6% | — |
| LMCA | 9.8% | — |
| Epoch Capabilities Index | 127.25 | — |
| ForecastBench | 59.4 | — |

## Math

- GPT-4 Turbo: 8.6 (#316)
- Step 3.5 Flash: 40.6 (#77)

| Benchmark | GPT-4 Turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1272 | 1408 |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| MathArena Final-Answer Competitions | — | 66.8% |
| OTIS Mock AIME 2024-2025 | 6.7% | — |
| MATH Level 5 | 46.7% | — |

## Knowledge

- GPT-4 Turbo: 24.3 (#263)
- Step 3.5 Flash: 39.0 (#127)

| Benchmark | GPT-4 Turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1223 | 1422 |
| GPQA Diamond | 46.6% | — |
| Confabulations | 28.4% | — |
| MMLU | 81.3% | — |

## Multimodal

- GPT-4 Turbo: 30.0 (#104)
- Step 3.5 Flash: —

| Benchmark | GPT-4 Turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Vision | 1090 | — |

## Multilingual

- GPT-4 Turbo: 40.7 (#211)
- Step 3.5 Flash: 50.7 (#115)

| Benchmark | GPT-4 Turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1245 | 1386 |
| LMArena Chinese | 1242 | 1448 |
| LMArena French | 1275 | 1421 |
| LMArena German | 1259 | 1404 |
| LMArena Japanese | 1194 | 1353 |
| LMArena Korean | 1187 | 1350 |
| LMArena Russian | 1259 | 1385 |
| LMArena Spanish | 1260 | 1419 |

## Instruction Following

- GPT-4 Turbo: 66.0 (#211)
- Step 3.5 Flash: 73.3 (#119)

| Benchmark | GPT-4 Turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1250 | 1385 |

## Long Context

- GPT-4 Turbo: 38.0 (#200)
- Step 3.5 Flash: 42.8 (#112)

| Benchmark | GPT-4 Turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1254 | 1402 |

## Writing & Preference

- GPT-4 Turbo: 47.9 (#201)
- Step 3.5 Flash: 59.0 (#108)

| Benchmark | GPT-4 Turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1272 | 1403 |
| LMArena Creative Writing | 1269 | 1357 |
| LMArena Multi-Turn | 1267 | 1405 |

## FAQ

### Is GPT-4 Turbo better than Step 3.5 Flash?

Step 3.5 Flash is the stronger model overall, scoring 41.7 to 30.3 on the Noometry Index.

### Which is cheaper, GPT-4 Turbo or Step 3.5 Flash?

Step 3.5 Flash is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; GPT-4 Turbo lists at $10 and $30.

### Is GPT-4 Turbo or Step 3.5 Flash better for coding?

Step 3.5 Flash scores higher on coding benchmarks: 41.7 versus 33.4 in the Noometry coding category.

### Which has the bigger context window?

Step 3.5 Flash does, with 256K tokens against 128K.

### How many benchmarks do GPT-4 Turbo and Step 3.5 Flash share?

17 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Step 3.5 Flash has 19.