# Qwen3.5 397B-A17B vs Step 3

> Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 39.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen3-5-397b-a17b-vs-step-3
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Qwen3.5 397B-A17B scores higher in 9 categories and Step 3 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5 397B-A17B leads 53.3 to 36.2.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 73.7% for Qwen3.5 397B-A17B and 62.3% for Step 3.

## Snapshot

| | Qwen3.5 397B-A17B | Step 3 |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 45.7 | 39.9 |
| Rank | 60 | 145 |
| Context | 262K | — |
| Input $/M | $0.60 | — |
| Output $/M | $3.60 | — |
| Weights | Open | Open |

## Coding

- Qwen3.5 397B-A17B: 41.7 (#108)
- Step 3: 39.5 (#144)

| Benchmark | Qwen3.5 397B-A17B | Step 3 |
|---|---|---|
| LMArena Coding | 1464 | 1367 |
| LMArena WebDev | 1400 | — |

## Agentic & Tool Use

- Qwen3.5 397B-A17B: 33.2 (#51)
- Step 3: —

| Benchmark | Qwen3.5 397B-A17B | Step 3 |
|---|---|---|
| APEX-Agents | 24.9% | — |
| τ²-bench Airline | 81.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 84.4% | — |
| τ²-bench Telecom | 97.8% | — |

## Reasoning

- Qwen3.5 397B-A17B: 33.6 (#62)
- Step 3: 27.6 (#102)

| Benchmark | Qwen3.5 397B-A17B | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 73.7% | 62.3% |
| LMArena Hard Prompts | 1448 | 1355 |
| NYT Connections (extended) | 58.9% | — |
| Chess Puzzles | 13% | — |
| Thematic Generalization | 65.1% | — |
| Mystery Game Puzzles | 18% | — |
| DTBench | 87.5% | — |
| LMCA | 37.9% | — |
| Epoch Capabilities Index | 146.64 | — |

## Math

- Qwen3.5 397B-A17B: 44.9 (#64)
- Step 3: 34.9 (#148)

| Benchmark | Qwen3.5 397B-A17B | Step 3 |
|---|---|---|
| LMArena Math | 1455 | 1366 |
| FrontierMath (Tiers 1-3) | 31.2% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |

## Knowledge

- Qwen3.5 397B-A17B: 53.3 (#53)
- Step 3: 36.2 (#161)

| Benchmark | Qwen3.5 397B-A17B | Step 3 |
|---|---|---|
| LMArena Expert | 1461 | 1333 |
| GPQA Diamond | 86.4% | — |

## Multimodal

- Qwen3.5 397B-A17B: 40.0 (#42)
- Step 3: 34.9 (#81)

| Benchmark | Qwen3.5 397B-A17B | Step 3 |
|---|---|---|
| LMArena Vision | 1262 | 1176 |

## Multilingual

- Qwen3.5 397B-A17B: 54.0 (#56)
- Step 3: 46.5 (#154)

| Benchmark | Qwen3.5 397B-A17B | Step 3 |
|---|---|---|
| LMArena Non-English | 1431 | 1327 |
| LMArena Chinese | 1501 | 1396 |
| LMArena German | 1447 | 1371 |
| LMArena Korean | 1382 | 1271 |
| LMArena Russian | 1429 | 1331 |
| LMArena Spanish | 1443 | 1372 |
| LMArena French | 1461 | — |
| LMArena Japanese | 1427 | — |

## Instruction Following

- Qwen3.5 397B-A17B: 75.2 (#73)
- Step 3: 70.6 (#159)

| Benchmark | Qwen3.5 397B-A17B | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1424 | 1332 |

## Long Context

- Qwen3.5 397B-A17B: 44.1 (#71)
- Step 3: 40.3 (#152)

| Benchmark | Qwen3.5 397B-A17B | Step 3 |
|---|---|---|
| LMArena Longer Query | 1442 | 1326 |

## Writing & Preference

- Qwen3.5 397B-A17B: 62.5 (#74)
- Step 3: 54.5 (#146)

| Benchmark | Qwen3.5 397B-A17B | Step 3 |
|---|---|---|
| LMArena Text | 1438 | 1351 |
| LMArena Creative Writing | 1401 | 1322 |
| LMArena Multi-Turn | 1447 | 1340 |
| EQ-Bench Creative Writing | 1478 | — |

## FAQ

### Is Qwen3.5 397B-A17B better than Step 3?

Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 39.9 on the Noometry Index.

### Is Qwen3.5 397B-A17B or Step 3 better for coding?

Qwen3.5 397B-A17B scores higher on coding benchmarks: 41.7 versus 39.5 in the Noometry coding category.

### How many benchmarks do Qwen3.5 397B-A17B and Step 3 share?

17 benchmarks have published results for both models. Qwen3.5 397B-A17B has 36 scored results on Noometry and Step 3 has 17.