# GPT-5.5 vs Step 3

> GPT-5.5 is the stronger model overall, scoring 63.4 to 39.9 on the Noometry Index.

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

## Summary

- They share 17 benchmarks with published results for both. GPT-5.5 scores higher in 9 categories and Step 3 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.5 leads 81.5 to 34.9.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 88.8% for GPT-5.5 and 62.3% for Step 3.
- Step 3 has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-5.5 | Step 3 |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 63.4 | 39.9 |
| Rank | 8 | 145 |
| Context | 1.05M | — |
| Input $/M | $5 | — |
| Output $/M | $30 | — |
| Weights | Proprietary | Open |

## Coding

- GPT-5.5: 58.3 (#13)
- Step 3: 39.5 (#144)

| Benchmark | GPT-5.5 | Step 3 |
|---|---|---|
| LMArena Coding | 1494 | 1367 |
| SWE-bench Verified | 80.6% | — |
| DeepSWE | 67% | — |
| FrontierCode | 43% | — |
| LMArena WebDev | 1512 | — |
| SciCode | 56.1% | — |
| GSO | 40.2% | — |
| WeirdML | 84.9% | — |
| MirrorCode | 10% | — |
| ALE-Bench | 1,943 | — |

## Agentic & Tool Use

- GPT-5.5: 50.7 (#6)
- Step 3: —

| Benchmark | GPT-5.5 | Step 3 |
|---|---|---|
| Terminal-Bench | 84.7% | — |
| APEX-Agents | 55.1% | — |
| OSWorld 2.0 | 13% | — |
| Remote Labor Index | 6.3% | — |
| τ²-bench Banking | 44.6% | — |
| DeepResearch Bench | 54% | — |
| PostTrainBench | 27.2% | — |
| ExploitBench | 47.4% | — |
| GBAEval | 53.2% | — |
| GDP.pdf | 26% | — |
| LMArena Search | 1242 | — |
| Vending-Bench 2 | 7,524 | — |

## Reasoning

- GPT-5.5: 72.7 (#8)
- Step 3: 27.6 (#102)

| Benchmark | GPT-5.5 | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 88.8% | 62.3% |
| LMArena Hard Prompts | 1489 | 1355 |
| ARC-AGI-2 | 85% | — |
| SimpleBench | 69% | — |
| NYT Connections (extended) | 96.2% | — |
| ARC-AGI-1 | 95% | — |
| CritPt | 27.1% | — |
| Chess Puzzles | 54% | — |
| EBR-Bench | 34.3% | — |
| Mystery Game Puzzles | 56% | — |
| DTBench | 96% | — |
| LMCA | 54.3% | — |
| Surface Evolver Bench | 88.1% | — |
| Bench to the Future 3 | 0.14 | — |
| Epoch Capabilities Index | 159.25 | — |
| ForecastBench | 60.6 | — |

## Math

- GPT-5.5: 81.5 (#9)
- Step 3: 34.9 (#148)

| Benchmark | GPT-5.5 | Step 3 |
|---|---|---|
| LMArena Math | 1486 | 1366 |
| FrontierMath (Tiers 1-3) | 85.3% | — |
| FrontierMath Tier 4 | 72.5% | — |
| MathArena Final-Answer Competitions | 94.3% | — |
| OTIS Mock AIME 2024-2025 | 100% | — |
| ProofBench | 50% | — |
| FrontierMath (Feb 2025 set) | 51.7% | — |
| FrontierMath Erdős | 0% | — |
| FrontierMath Tier 4 (v1) | 35.4% | — |

## Knowledge

- GPT-5.5: 64.4 (#14)
- Step 3: 36.2 (#161)

| Benchmark | GPT-5.5 | Step 3 |
|---|---|---|
| LMArena Expert | 1508 | 1333 |
| GPQA Diamond | 94% | — |
| SimpleQA Verified | 63% | — |
| Vectara Hallucination Rate | 9.3% | — |

## Multimodal

- GPT-5.5: 47.1 (#11)
- Step 3: 34.9 (#81)

| Benchmark | GPT-5.5 | Step 3 |
|---|---|---|
| LMArena Vision | 1297 | 1176 |
| Blueprint-Bench 2 | 36.2% | — |
| Furniture Assembly | 44.2% | — |
| LMArena Document | 1486 | — |

## Multilingual

- GPT-5.5: 56.5 (#19)
- Step 3: 46.5 (#154)

| Benchmark | GPT-5.5 | Step 3 |
|---|---|---|
| LMArena Non-English | 1467 | 1327 |
| LMArena Chinese | 1535 | 1396 |
| LMArena German | 1478 | 1371 |
| LMArena Korean | 1461 | 1271 |
| LMArena Russian | 1473 | 1331 |
| LMArena Spanish | 1469 | 1372 |
| LMArena French | 1486 | — |
| LMArena Japanese | 1498 | — |

## Instruction Following

- GPT-5.5: 77.7 (#18)
- Step 3: 70.6 (#159)

| Benchmark | GPT-5.5 | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1479 | 1332 |

## Long Context

- GPT-5.5: 48.3 (#12)
- Step 3: 40.3 (#152)

| Benchmark | GPT-5.5 | Step 3 |
|---|---|---|
| LMArena Longer Query | 1484 | 1326 |
| CL-bench Life | 22.2% | — |

## Writing & Preference

- GPT-5.5: 72.8 (#13)
- Step 3: 54.5 (#146)

| Benchmark | GPT-5.5 | Step 3 |
|---|---|---|
| LMArena Text | 1471 | 1351 |
| LMArena Creative Writing | 1455 | 1322 |
| LMArena Multi-Turn | 1477 | 1340 |
| EQ-Bench Creative Writing | 1844 | — |
| EQ-Bench 4 | 1315 | — |

## FAQ

### Is GPT-5.5 better than Step 3?

GPT-5.5 is the stronger model overall, scoring 63.4 to 39.9 on the Noometry Index.

### Is GPT-5.5 or Step 3 better for coding?

GPT-5.5 scores higher on coding benchmarks: 58.3 versus 39.5 in the Noometry coding category.

### How many benchmarks do GPT-5.5 and Step 3 share?

17 benchmarks have published results for both models. GPT-5.5 has 71 scored results on Noometry and Step 3 has 17.