# Grok 4 vs Qwen1.5-72B

> Grok 4 is the stronger model overall, scoring 47.8 to 30.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-4-vs-qwen1-5-72b
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. Grok 4 scores higher in 8 categories and Qwen1.5-72B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok 4 leads 53.7 to 11.5.
- The biggest single-benchmark swing is GPQA Diamond: 87% for Grok 4 and 28.8% for Qwen1.5-72B.
- Qwen1.5-72B has downloadable open weights; the other is API-only.

## Snapshot

| | Grok 4 | Qwen1.5-72B |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 47.8 | 30.3 |
| Rank | 49 | 282 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Proprietary | Open |

## Coding

- Grok 4: 49.9 (#44)
- Qwen1.5-72B: 31.4 (#272)

| Benchmark | Grok 4 | Qwen1.5-72B |
|---|---|---|
| LMArena Coding | 1408 | 1166 |
| Aider Polyglot | 79.6% | — |
| WeirdML | 45.7% | — |
| BigCodeBench Instruct | — | 33.2% |
| BigCodeBench Complete | — | 40.3% |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 61.6% |

## Agentic & Tool Use

- Grok 4: 32.1 (#62)
- Qwen1.5-72B: —

| Benchmark | Grok 4 | Qwen1.5-72B |
|---|---|---|
| Terminal-Bench | 27.2% | — |
| Berkeley Function Calling Leaderboard | 63% | — |
| GDPval | 21.1% | — |
| Cybench | 43% | — |
| DeepResearch Bench | 47.3% | — |
| BALROG | 43.6% | — |
| LMArena Search | 1142 | — |
| METR Time Horizons | 66.6% | — |

## Reasoning

- Grok 4: 36.9 (#59)
- Qwen1.5-72B: 21.2 (#201)

| Benchmark | Grok 4 | Qwen1.5-72B |
|---|---|---|
| LMArena Hard Prompts | 1409 | 1148 |
| ARC-AGI-2 | 16% | — |
| SimpleBench | 60.5% | — |
| Kagi LLM Benchmark | 73.6% | — |
| ARC-AGI-1 | 66.7% | — |
| Chess Puzzles | 28% | — |
| Epoch Capabilities Index | 146.46 | — |
| ForecastBench | 60.9 | — |

## Math

- Grok 4: 46.5 (#58)
- Qwen1.5-72B: 30.7 (#205)

| Benchmark | Grok 4 | Qwen1.5-72B |
|---|---|---|
| LMArena Math | 1422 | 1164 |
| OTIS Mock AIME 2024-2025 | 84% | — |
| Omni-MATH | 60.3% | — |
| FrontierMath (Feb 2025 set) | 19.7% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- Grok 4: 53.7 (#50)
- Qwen1.5-72B: 11.5 (#295)

| Benchmark | Grok 4 | Qwen1.5-72B |
|---|---|---|
| GPQA Diamond | 87% | 28.8% |
| LMArena Expert | 1417 | 1136 |
| MMLU-Pro | 85.1% | — |
| Confabulations | 12.4% | — |
| GPQA (HELM) | 72.7% | — |

## Multimodal

- Grok 4: 33.1 (#89)
- Qwen1.5-72B: —

| Benchmark | Grok 4 | Qwen1.5-72B |
|---|---|---|
| LMArena Vision | 1211 | — |
| GeoBench | 45% | — |

## Multilingual

- Grok 4: 52.0 (#98)
- Qwen1.5-72B: 33.4 (#248)

| Benchmark | Grok 4 | Qwen1.5-72B |
|---|---|---|
| LMArena Non-English | 1403 | 1135 |
| LMArena Chinese | 1428 | 1186 |
| LMArena French | 1418 | 1158 |
| LMArena German | 1429 | 1084 |
| LMArena Japanese | 1396 | 1061 |
| LMArena Korean | 1376 | 1049 |
| LMArena Russian | 1409 | 1104 |
| LMArena Spanish | 1419 | 1110 |

## Instruction Following

- Grok 4: 79.4 (#4)
- Qwen1.5-72B: 59.6 (#251)

| Benchmark | Grok 4 | Qwen1.5-72B |
|---|---|---|
| LMArena Instruction Following | 1387 | 1142 |
| IFEval | 94.9% | — |

## Long Context

- Grok 4: 63.1 (#4)
- Qwen1.5-72B: 35.0 (#238)

| Benchmark | Grok 4 | Qwen1.5-72B |
|---|---|---|
| LMArena Longer Query | 1409 | 1157 |
| Fiction.LiveBench | 94.4% | — |

## Writing & Preference

- Grok 4: 58.7 (#111)
- Qwen1.5-72B: 37.4 (#253)

| Benchmark | Grok 4 | Qwen1.5-72B |
|---|---|---|
| LMArena Text | 1412 | 1166 |
| LMArena Creative Writing | 1397 | 1137 |
| LMArena Multi-Turn | 1416 | 1160 |
| Short-Story Creative Writing | 76.9% | — |
| WildBench | 79.7% | — |

## FAQ

### Is Grok 4 better than Qwen1.5-72B?

Grok 4 is the stronger model overall, scoring 47.8 to 30.3 on the Noometry Index.

### Is Grok 4 or Qwen1.5-72B better for coding?

Grok 4 scores higher on coding benchmarks: 49.9 versus 31.4 in the Noometry coding category.

### How many benchmarks do Grok 4 and Qwen1.5-72B share?

18 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Qwen1.5-72B has 22.