# DeepSeek V4 Pro vs Qwen3.5 Max Preview

> DeepSeek V4 Pro is the stronger model overall, scoring 54.2 to 44.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v4-pro-vs-qwen3-5-max-preview
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 4 categories and Qwen3.5 Max Preview in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Pro leads 64.6 to 37.3.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek V4 Pro | Qwen3.5 Max Preview |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 54.2 | 44.7 |
| Rank | 28 | 72 |
| Context | 1M | — |
| Input $/M | $0.43 | — |
| Output $/M | $0.87 | — |
| Weights | Open | Proprietary |

## Coding

- DeepSeek V4 Pro: 52.2 (#31)
- Qwen3.5 Max Preview: 43.4 (#79)

| Benchmark | DeepSeek V4 Pro | Qwen3.5 Max Preview |
|---|---|---|
| LMArena Coding | 1470 | 1488 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| LMArena WebDev | 1582 | — |
| SciCode | 51% | — |
| WeirdML | 66.2% | — |
| ALE-Bench | 1,403 | — |

## Agentic & Tool Use

- DeepSeek V4 Pro: 32.7 (#54)
- Qwen3.5 Max Preview: —

| Benchmark | DeepSeek V4 Pro | Qwen3.5 Max Preview |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Vending-Bench 2 | 3,285 | — |

## Reasoning

- DeepSeek V4 Pro: 56.4 (#22)
- Qwen3.5 Max Preview: 29.6 (#81)

| Benchmark | DeepSeek V4 Pro | Qwen3.5 Max Preview |
|---|---|---|
| LMArena Hard Prompts | 1461 | 1483 |
| ARC-AGI-2 | 61.3% | — |
| Kagi LLM Benchmark | 53.5% | — |
| NYT Connections (extended) | 91.3% | — |
| ARC-AGI-1 | 90.5% | — |
| CritPt | 18% | — |
| Chess Puzzles | 47% | — |
| Mystery Game Puzzles | 43% | — |
| DTBench | 93.9% | — |
| LMCA | 45.5% | — |
| Surface Evolver Bench | 40% | — |
| Epoch Capabilities Index | 155.39 | — |
| ForecastBench | 56.1 | — |

## Math

- DeepSeek V4 Pro: 64.6 (#24)
- Qwen3.5 Max Preview: 37.3 (#99)

| Benchmark | DeepSeek V4 Pro | Qwen3.5 Max Preview |
|---|---|---|
| LMArena Math | 1455 | 1473 |
| FrontierMath (Tiers 1-3) | 64.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| OTIS Mock AIME 2024-2025 | 98.6% | — |
| ProofBench | 50% | — |

## Knowledge

- DeepSeek V4 Pro: 59.5 (#28)
- Qwen3.5 Max Preview: 41.1 (#104)

| Benchmark | DeepSeek V4 Pro | Qwen3.5 Max Preview |
|---|---|---|
| LMArena Expert | 1463 | 1489 |
| GPQA Diamond | 91.7% | — |
| SimpleQA Verified | 52.9% | — |
| Vectara Hallucination Rate | 8.6% | — |

## Multilingual

- DeepSeek V4 Pro: 54.5 (#44)
- Qwen3.5 Max Preview: 56.4 (#21)

| Benchmark | DeepSeek V4 Pro | Qwen3.5 Max Preview |
|---|---|---|
| LMArena Non-English | 1438 | 1465 |
| LMArena Chinese | 1486 | 1534 |
| LMArena French | 1473 | 1483 |
| LMArena German | 1458 | 1487 |
| LMArena Japanese | 1448 | 1497 |
| LMArena Korean | 1452 | 1438 |
| LMArena Russian | 1452 | 1472 |
| LMArena Spanish | 1457 | 1470 |

## Instruction Following

- DeepSeek V4 Pro: 76.3 (#44)
- Qwen3.5 Max Preview: 77.2 (#28)

| Benchmark | DeepSeek V4 Pro | Qwen3.5 Max Preview |
|---|---|---|
| LMArena Instruction Following | 1448 | 1467 |

## Long Context

- DeepSeek V4 Pro: 45.0 (#49)
- Qwen3.5 Max Preview: 45.2 (#44)

| Benchmark | DeepSeek V4 Pro | Qwen3.5 Max Preview |
|---|---|---|
| LMArena Longer Query | 1459 | 1477 |
| CL-bench Life | 13.5% | — |

## Writing & Preference

- DeepSeek V4 Pro: 65.6 (#45)
- Qwen3.5 Max Preview: 66.1 (#39)

| Benchmark | DeepSeek V4 Pro | Qwen3.5 Max Preview |
|---|---|---|
| LMArena Text | 1451 | 1471 |
| LMArena Creative Writing | 1446 | 1464 |
| LMArena Multi-Turn | 1467 | 1478 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |

## FAQ

### Is DeepSeek V4 Pro better than Qwen3.5 Max Preview?

DeepSeek V4 Pro is the stronger model overall, scoring 54.2 to 44.7 on the Noometry Index.

### Is DeepSeek V4 Pro or Qwen3.5 Max Preview better for coding?

DeepSeek V4 Pro scores higher on coding benchmarks: 52.2 versus 43.4 in the Noometry coding category.

### How many benchmarks do DeepSeek V4 Pro and Qwen3.5 Max Preview share?

17 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Qwen3.5 Max Preview has 17.