# Grok-2 (Dec 2024) vs Qwen3 Max

> Qwen3 Max is the stronger model overall, scoring 43.7 to 33.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-2-vs-qwen3-max
- Last updated: 2026-10-11
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. Grok-2 (Dec 2024) scores higher in 0 categories and Qwen3 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 Max leads 48.1 to 29.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 11.5% for Grok-2 (Dec 2024) and 73.3% for Qwen3 Max.

## Snapshot

| | Grok-2 (Dec 2024) | Qwen3 Max |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 33.7 | 43.7 |
| Rank | 239 | 87 |
| Context | — | 262K |
| Input $/M | — | $1.20 |
| Output $/M | — | $6 |
| Weights | Proprietary | Proprietary |

## Coding

- Grok-2 (Dec 2024): 33.3 (#258)
- Qwen3 Max: 43.0 (#93)

| Benchmark | Grok-2 (Dec 2024) | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1287 | 1456 |
| WeirdML | 22.2% | — |
| LiveBench Coding | 46.4% | — |
| ALE-Bench | — | 370.45 |

## Agentic & Tool Use

- Grok-2 (Dec 2024): —
- Qwen3 Max: —

| Benchmark | Grok-2 (Dec 2024) | Qwen3 Max |
|---|---|---|
| Vending-Bench 2 | — | 71.56 |

## Reasoning

- Grok-2 (Dec 2024): 16.9 (#299)
- Qwen3 Max: 22.6 (#190)

| Benchmark | Grok-2 (Dec 2024) | Qwen3 Max |
|---|---|---|
| LMArena Hard Prompts | 1272 | 1448 |
| DTBench | 65.2% | 82.1% |
| Epoch Capabilities Index | 130.48 | 142.38 |
| SimpleBench | 22.7% | — |
| Kagi LLM Benchmark | — | 72.5% |
| NYT Connections (extended) | — | 30.1% |
| Chess Puzzles | — | 4% |
| LiveBench Reasoning | 54.8% | — |
| Mystery Game Puzzles | — | 5% |
| LiveBench Data Analysis | 54.5% | — |
| LMCA | — | 28.3% |
| LiveBench | 54.3% | — |

## Math

- Grok-2 (Dec 2024): 20.8 (#284)
- Qwen3 Max: 38.7 (#131)

| Benchmark | Grok-2 (Dec 2024) | Qwen3 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 11.5% | 73.3% |
| LMArena Math | 1283 | 1446 |
| MATH Level 5 | 63.5% | 97.1% |
| FrontierMath (Tiers 1-3) | — | 18.9% |
| LiveBench Math | 54.9% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |

## Knowledge

- Grok-2 (Dec 2024): 29.8 (#233)
- Qwen3 Max: 48.1 (#78)

| Benchmark | Grok-2 (Dec 2024) | Qwen3 Max |
|---|---|---|
| GPQA Diamond | 53.8% | 72.6% |
| LMArena Expert | 1254 | 1455 |
| SimpleQA Verified | — | 48.7% |
| Confabulations | 20.1% | — |

## Multilingual

- Grok-2 (Dec 2024): 43.1 (#188)
- Qwen3 Max: 53.7 (#62)

| Benchmark | Grok-2 (Dec 2024) | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1282 | 1429 |
| LMArena Chinese | 1289 | 1478 |
| LMArena French | 1318 | 1449 |
| LMArena German | 1287 | 1463 |
| LMArena Japanese | 1244 | 1397 |
| LMArena Korean | 1237 | 1399 |
| LMArena Russian | 1286 | 1428 |
| LMArena Spanish | 1281 | 1462 |

## Instruction Following

- Grok-2 (Dec 2024): 66.9 (#202)
- Qwen3 Max: 74.8 (#87)

| Benchmark | Grok-2 (Dec 2024) | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1270 | 1419 |
| LiveBench Instruction Following | 69.6% | — |

## Long Context

- Grok-2 (Dec 2024): 38.8 (#190)
- Qwen3 Max: 41.6 (#134)

| Benchmark | Grok-2 (Dec 2024) | Qwen3 Max |
|---|---|---|
| LMArena Longer Query | 1276 | 1438 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 14.5% |

## Writing & Preference

- Grok-2 (Dec 2024): 48.6 (#198)
- Qwen3 Max: 62.4 (#76)

| Benchmark | Grok-2 (Dec 2024) | Qwen3 Max |
|---|---|---|
| LMArena Text | 1305 | 1439 |
| LMArena Creative Writing | 1284 | 1402 |
| LMArena Multi-Turn | 1290 | 1446 |
| Short-Story Creative Writing | 63.6% | — |
| LiveBench Language | 45.6% | — |

## FAQ

### Is Grok-2 (Dec 2024) better than Qwen3 Max?

Qwen3 Max is the stronger model overall, scoring 43.7 to 33.7 on the Noometry Index.

### Is Grok-2 (Dec 2024) or Qwen3 Max better for coding?

Qwen3 Max scores higher on coding benchmarks: 43.0 versus 33.3 in the Noometry coding category.

### How many benchmarks do Grok-2 (Dec 2024) and Qwen3 Max share?

22 benchmarks have published results for both models. Grok-2 (Dec 2024) has 34 scored results on Noometry and Qwen3 Max has 33.