# DeepSeek-V2.5 (Sep 2024) vs Qwen3 Max

> Qwen3 Max is the stronger model overall, scoring 43.7 to 37.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v2-5-vs-qwen3-max
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 1 category and Qwen3 Max in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 Max leads 48.1 to 34.8.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek-V2.5 (Sep 2024) | Qwen3 Max |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 37.6 | 43.7 |
| Rank | 200 | 87 |
| Context | — | 262K |
| Input $/M | — | $1.20 |
| Output $/M | — | $6 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek-V2.5 (Sep 2024): 31.7 (#281)
- Qwen3 Max: 43.0 (#93)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1309 | 1456 |
| Aider Polyglot | 17.8% | — |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| ALE-Bench | — | 370.45 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |

## Agentic & Tool Use

- DeepSeek-V2.5 (Sep 2024): —
- Qwen3 Max: —

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3 Max |
|---|---|---|
| Vending-Bench 2 | — | 71.56 |

## Reasoning

- DeepSeek-V2.5 (Sep 2024): 25.6 (#145)
- Qwen3 Max: 22.6 (#190)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3 Max |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1448 |
| Kagi LLM Benchmark | — | 72.5% |
| NYT Connections (extended) | — | 30.1% |
| Chess Puzzles | — | 4% |
| Mystery Game Puzzles | — | 5% |
| DTBench | — | 82.1% |
| LMCA | — | 28.3% |
| Epoch Capabilities Index | — | 142.38 |

## Math

- DeepSeek-V2.5 (Sep 2024): 35.9 (#177)
- Qwen3 Max: 38.7 (#131)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3 Max |
|---|---|---|
| LMArena Math | 1288 | 1446 |
| FrontierMath (Tiers 1-3) | — | 18.9% |
| OTIS Mock AIME 2024-2025 | — | 73.3% |
| MATH Level 5 | — | 97.1% |

## Knowledge

- DeepSeek-V2.5 (Sep 2024): 34.8 (#193)
- Qwen3 Max: 48.1 (#78)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3 Max |
|---|---|---|
| LMArena Expert | 1266 | 1455 |
| GPQA Diamond | — | 72.6% |
| SimpleQA Verified | — | 48.7% |

## Multilingual

- DeepSeek-V2.5 (Sep 2024): 42.5 (#193)
- Qwen3 Max: 53.7 (#62)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1273 | 1429 |
| LMArena Chinese | 1318 | 1478 |
| LMArena French | 1289 | 1449 |
| LMArena German | 1258 | 1463 |
| LMArena Japanese | 1228 | 1397 |
| LMArena Korean | 1209 | 1399 |
| LMArena Russian | 1289 | 1428 |
| LMArena Spanish | 1248 | 1462 |

## Instruction Following

- DeepSeek-V2.5 (Sep 2024): 67.5 (#194)
- Qwen3 Max: 74.8 (#87)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1280 | 1419 |

## Long Context

- DeepSeek-V2.5 (Sep 2024): 39.5 (#174)
- Qwen3 Max: 41.6 (#134)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3 Max |
|---|---|---|
| LMArena Longer Query | 1301 | 1438 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 14.5% |

## Writing & Preference

- DeepSeek-V2.5 (Sep 2024): 49.8 (#187)
- Qwen3 Max: 62.4 (#76)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3 Max |
|---|---|---|
| LMArena Text | 1294 | 1439 |
| LMArena Creative Writing | 1285 | 1402 |
| LMArena Multi-Turn | 1297 | 1446 |

## FAQ

### Is DeepSeek-V2.5 (Sep 2024) better than Qwen3 Max?

Qwen3 Max is the stronger model overall, scoring 43.7 to 37.6 on the Noometry Index.

### Is DeepSeek-V2.5 (Sep 2024) or Qwen3 Max better for coding?

Qwen3 Max scores higher on coding benchmarks: 43.0 versus 31.7 in the Noometry coding category.

### How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Qwen3 Max share?

17 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Qwen3 Max has 33.