# Qwen3.8 Max vs WizardLM-2 8x22B

> Qwen3.8 Max has enough public results to be ranked (#21); WizardLM-2 8x22B does not yet, so treat this comparison as directional.

- Canonical page: https://noometry.com/compare/qwen3-8-max-vs-wizardlm-2-8x22b
- Last updated: 2026-10-10
- Shared benchmarks: 1

## Summary

- They share 1 benchmark with published results for both. Qwen3.8 Max scores higher in 2 categories and WizardLM-2 8x22B in 0 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.1 to 26.8.
- The biggest single-benchmark swing is GPQA Diamond: 92.7% for Qwen3.8 Max and 43.4% for WizardLM-2 8x22B.
- WizardLM-2 8x22B is cheaper at $0.62 / $0.62 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 Max accepts more context: 1M tokens versus 66K.
- WizardLM-2 8x22B has downloadable open weights; the other is API-only.

## Snapshot

| | Qwen3.8 Max | WizardLM-2 8x22B |
|---|---|---|
| Provider | Alibaba (Qwen) | Microsoft |
| Noometry Index | 56.0 | 30.7 |
| Rank | 21 | — |
| Context | 1M | 66K |
| Input $/M | $2 | $0.62 |
| Output $/M | $6 | $0.62 |
| Weights | Proprietary | Open |

## Coding

- Qwen3.8 Max: 52.3 (#30)
- WizardLM-2 8x22B: —

| Benchmark | Qwen3.8 Max | WizardLM-2 8x22B |
|---|---|---|
| DeepSWE | 57.5% | — |
| LMArena WebDev | 1671 | — |
| FrontierSWE | 15.8% | — |
| SciCode | 52.9% | — |
| LMArena Coding | 1499 | — |

## Agentic & Tool Use

- Qwen3.8 Max: 42.6 (#17)
- WizardLM-2 8x22B: —

| Benchmark | Qwen3.8 Max | WizardLM-2 8x22B |
|---|---|---|
| τ²-bench Banking | 55.1% | — |
| GDP.pdf | 23.2% | — |

## Reasoning

- Qwen3.8 Max: 53.4 (#26)
- WizardLM-2 8x22B: —

| Benchmark | Qwen3.8 Max | WizardLM-2 8x22B |
|---|---|---|
| NYT Connections (extended) | 88.3% | — |
| CritPt | 20% | — |
| Chess Puzzles | 40% | — |
| LMArena Hard Prompts | 1493 | — |
| Mystery Game Puzzles | 38% | — |
| DTBench | 92% | — |
| LMCA | 46.2% | — |
| Epoch Capabilities Index | 156.69 | — |

## Math

- Qwen3.8 Max: 73.1 (#15)
- WizardLM-2 8x22B: 26.8

| Benchmark | Qwen3.8 Max | WizardLM-2 8x22B |
|---|---|---|
| FrontierMath (Tiers 1-3) | 74.7% | — |
| FrontierMath Tier 4 | 46.3% | — |
| OTIS Mock AIME 2024-2025 | 100% | — |
| ProofBench | 58% | — |
| LMArena Math | 1498 | — |
| MATH Level 5 | — | 25.7% |

## Knowledge

- Qwen3.8 Max: 61.7 (#25)
- WizardLM-2 8x22B: 23.3

| Benchmark | Qwen3.8 Max | WizardLM-2 8x22B |
|---|---|---|
| GPQA Diamond | 92.7% | 43.4% |
| SimpleQA Verified | 47.3% | — |
| LMArena Expert | 1504 | — |

## Multimodal

- Qwen3.8 Max: 36.9 (#70)
- WizardLM-2 8x22B: —

| Benchmark | Qwen3.8 Max | WizardLM-2 8x22B |
|---|---|---|
| LMArena Vision | 1314 | — |
| Furniture Assembly | 20% | — |

## Multilingual

- Qwen3.8 Max: 56.6 (#18)
- WizardLM-2 8x22B: —

| Benchmark | Qwen3.8 Max | WizardLM-2 8x22B |
|---|---|---|
| LMArena Non-English | 1468 | — |
| LMArena Chinese | 1537 | — |
| LMArena French | 1502 | — |
| LMArena German | 1489 | — |
| LMArena Japanese | 1466 | — |
| LMArena Korean | 1467 | — |
| LMArena Russian | 1478 | — |
| LMArena Spanish | 1495 | — |

## Instruction Following

- Qwen3.8 Max: 77.7 (#19)
- WizardLM-2 8x22B: —

| Benchmark | Qwen3.8 Max | WizardLM-2 8x22B |
|---|---|---|
| LMArena Instruction Following | 1478 | — |

## Long Context

- Qwen3.8 Max: 45.5 (#30)
- WizardLM-2 8x22B: —

| Benchmark | Qwen3.8 Max | WizardLM-2 8x22B |
|---|---|---|
| LMArena Longer Query | 1486 | — |

## Writing & Preference

- Qwen3.8 Max: 67.0 (#31)
- WizardLM-2 8x22B: —

| Benchmark | Qwen3.8 Max | WizardLM-2 8x22B |
|---|---|---|
| LMArena Text | 1480 | — |
| LMArena Creative Writing | 1478 | — |
| LMArena Multi-Turn | 1487 | — |

## FAQ

### Is Qwen3.8 Max better than WizardLM-2 8x22B?

Qwen3.8 Max has enough public results to be ranked (#21); WizardLM-2 8x22B does not yet, so treat this comparison as directional.

### Which is cheaper, Qwen3.8 Max or WizardLM-2 8x22B?

WizardLM-2 8x22B is cheaper. It lists at $0.62 per million input tokens and $0.62 per million output tokens; Qwen3.8 Max lists at $2 and $6.

### Which has the bigger context window?

Qwen3.8 Max does, with 1M tokens against 66K.

### How many benchmarks do Qwen3.8 Max and WizardLM-2 8x22B share?

1 benchmark has published results for both models. Qwen3.8 Max has 38 scored results on Noometry and WizardLM-2 8x22B has 2.