# DeepSeek-V3 vs Qwen3.8 Max

> Qwen3.8 Max is the stronger model overall, scoring 56.0 to 39.2 on the Noometry Index. DeepSeek-V3 costs 7.4× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/deepseek-v3-vs-qwen3-8-max
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. DeepSeek-V3 scores higher in 0 categories and Qwen3.8 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.1 to 30.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for DeepSeek-V3 and 100% for Qwen3.8 Max.
- DeepSeek-V3 is cheaper at $0.24 / $0.90 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 Max accepts more context: 1M tokens versus 164K.
- DeepSeek-V3 has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek-V3 | Qwen3.8 Max |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 39.2 | 56.0 |
| Rank | 157 | 21 |
| Context | 164K | 1M |
| Input $/M | $0.24 | $2 |
| Output $/M | $0.90 | $6 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek-V3: 41.9 (#99)
- Qwen3.8 Max: 52.3 (#30)

| Benchmark | DeepSeek-V3 | Qwen3.8 Max |
|---|---|---|
| SciCode | 35.8% | 52.9% |
| LMArena Coding | 1368 | 1499 |
| DeepSWE | — | 57.5% |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1671 |
| FrontierSWE | — | 15.8% |
| WeirdML | 36.1% | — |
| BigCodeBench Instruct | 50% | — |
| LiveBench Coding | 70.9% | — |
| BigCodeBench Complete | 62.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73% | — |

## Agentic & Tool Use

- DeepSeek-V3: —
- Qwen3.8 Max: 42.6 (#17)

| Benchmark | DeepSeek-V3 | Qwen3.8 Max |
|---|---|---|
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
| METR Time Horizons | 49.6% | — |

## Reasoning

- DeepSeek-V3: 20.0 (#226)
- Qwen3.8 Max: 53.4 (#26)

| Benchmark | DeepSeek-V3 | Qwen3.8 Max |
|---|---|---|
| CritPt | 0% | 20% |
| LMArena Hard Prompts | 1365 | 1493 |
| DTBench | 64.8% | 92% |
| LMCA | 15.5% | 46.2% |
| Epoch Capabilities Index | 135.95 | 156.69 |
| SimpleBench | 27.2% | — |
| Kagi LLM Benchmark | 52.3% | — |
| NYT Connections (extended) | — | 88.3% |
| Chess Puzzles | — | 40% |
| LiveBench Reasoning | 65.8% | — |
| Mystery Game Puzzles | — | 38% |
| LiveBench Data Analysis | 60.9% | — |
| BIG-Bench Hard | 87.5% | — |
| ForecastBench | 59.1 | — |
| HellaSwag | 88.9% | — |
| LiveBench | 66.9% | — |
| PIQA | 84.7% | — |
| WinoGrande | 85.2% | — |

## Math

- DeepSeek-V3: 30.4 (#208)
- Qwen3.8 Max: 73.1 (#15)

| Benchmark | DeepSeek-V3 | Qwen3.8 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 100% |
| LMArena Math | 1373 | 1498 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| ProofBench | — | 58% |
| Omni-MATH | 40.3% | — |
| LiveBench Math | 73.5% | — |
| MATH Level 5 | 75.5% | — |
| FrontierMath (Feb 2025 set) | 1.7% | — |

## Knowledge

- DeepSeek-V3: 37.5 (#147)
- Qwen3.8 Max: 61.7 (#25)

| Benchmark | DeepSeek-V3 | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 67.6% | 92.7% |
| LMArena Expert | 1352 | 1504 |
| SimpleQA Verified | — | 47.3% |
| MMLU-Pro | 72.3% | — |
| Confabulations | 26.1% | — |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 53.8% | — |
| ARC (AI2) Challenge | 95.3% | — |
| MMLU | 87.2% | — |
| TriviaQA | 82.9% | — |

## Multimodal

- DeepSeek-V3: —
- Qwen3.8 Max: 36.9 (#70)

| Benchmark | DeepSeek-V3 | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |

## Multilingual

- DeepSeek-V3: 48.8 (#138)
- Qwen3.8 Max: 56.6 (#18)

| Benchmark | DeepSeek-V3 | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1358 | 1468 |
| LMArena Chinese | 1390 | 1537 |
| LMArena French | 1386 | 1502 |
| LMArena German | 1374 | 1489 |
| LMArena Japanese | 1333 | 1466 |
| LMArena Korean | 1321 | 1467 |
| LMArena Russian | 1372 | 1478 |
| LMArena Spanish | 1358 | 1495 |

## Instruction Following

- DeepSeek-V3: 73.0 (#125)
- Qwen3.8 Max: 77.7 (#19)

| Benchmark | DeepSeek-V3 | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1345 | 1478 |
| LiveBench Instruction Following | 81.5% | — |
| IFEval | 83.2% | — |

## Long Context

- DeepSeek-V3: 34.0 (#248)
- Qwen3.8 Max: 45.5 (#30)

| Benchmark | DeepSeek-V3 | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1353 | 1486 |
| Fiction.LiveBench | 50% | — |

## Writing & Preference

- DeepSeek-V3: 57.6 (#124)
- Qwen3.8 Max: 67.0 (#31)

| Benchmark | DeepSeek-V3 | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1375 | 1480 |
| LMArena Creative Writing | 1364 | 1478 |
| LMArena Multi-Turn | 1389 | 1487 |
| Short-Story Creative Writing | 77% | — |
| EQ-Bench Creative Writing | 1472 | — |
| WildBench | 83% | — |
| LiveBench Language | 49.1% | — |

## FAQ

### Is DeepSeek-V3 better than Qwen3.8 Max?

Qwen3.8 Max is the stronger model overall, scoring 56.0 to 39.2 on the Noometry Index. DeepSeek-V3 costs 7.4× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.

### Which is cheaper, DeepSeek-V3 or Qwen3.8 Max?

DeepSeek-V3 is cheaper. It lists at $0.24 per million input tokens and $0.90 per million output tokens; Qwen3.8 Max lists at $2 and $6.

### Is DeepSeek-V3 or Qwen3.8 Max better for coding?

Qwen3.8 Max scores higher on coding benchmarks: 52.3 versus 41.9 in the Noometry coding category.

### Which has the bigger context window?

Qwen3.8 Max does, with 1M tokens against 164K.

### How many benchmarks do DeepSeek-V3 and Qwen3.8 Max share?

24 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and Qwen3.8 Max has 38.