# Grok 4.20 Multi-Agent vs Qwen2.5 72B Instruct

> Grok 4.20 Multi-Agent is the stronger model overall, scoring 45.5 to 31.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-4-20-multi-agent-vs-qwen2-5-72b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Grok 4.20 Multi-Agent scores higher in 8 categories and Qwen2.5 72B Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 Multi-Agent leads 42.4 to 21.4.
- Grok 4.20 Multi-Agent is cheaper at $1.25 / $2.50 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- Grok 4.20 Multi-Agent accepts more context: 1M tokens versus 131K.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Grok 4.20 Multi-Agent | Qwen2.5 72B Instruct |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 45.5 | 31.5 |
| Rank | 62 | 259 |
| Context | 1M | 131K |
| Input $/M | $1.25 | $1.40 |
| Output $/M | $2.50 | $5.60 |
| Weights | Proprietary | Open |

## Coding

- Grok 4.20 Multi-Agent: 42.4 (#91)
- Qwen2.5 72B Instruct: 32.8 (#256)

| Benchmark | Grok 4.20 Multi-Agent | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Coding | 1458 | 1293 |
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |

## Agentic & Tool Use

- Grok 4.20 Multi-Agent: —
- Qwen2.5 72B Instruct: 22.0 (#124)

| Benchmark | Grok 4.20 Multi-Agent | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| LMArena Search | 1204 | — |
| METR Time Horizons | — | 35.8% |

## Reasoning

- Grok 4.20 Multi-Agent: 42.4 (#44)
- Qwen2.5 72B Instruct: 21.4 (#197)

| Benchmark | Grok 4.20 Multi-Agent | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1448 | 1271 |
| NYT Connections (extended) | 89.6% | — |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| Epoch Capabilities Index | — | 129.01 |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |

## Math

- Grok 4.20 Multi-Agent: 36.6 (#106)
- Qwen2.5 72B Instruct: 18.1 (#281)

| Benchmark | Grok 4.20 Multi-Agent | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Math | 1441 | 1283 |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| MATH Level 5 | — | 63.2% |

## Knowledge

- Grok 4.20 Multi-Agent: 39.7 (#116)
- Qwen2.5 72B Instruct: 26.9 (#247)

| Benchmark | Grok 4.20 Multi-Agent | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Expert | 1445 | 1245 |
| GPQA Diamond | — | 49.1% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |

## Multimodal

- Grok 4.20 Multi-Agent: 39.8 (#44)
- Qwen2.5 72B Instruct: —

| Benchmark | Grok 4.20 Multi-Agent | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1259 | — |

## Multilingual

- Grok 4.20 Multi-Agent: 54.6 (#40)
- Qwen2.5 72B Instruct: 41.2 (#208)

| Benchmark | Grok 4.20 Multi-Agent | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1440 | 1252 |
| LMArena Chinese | 1475 | 1272 |
| LMArena French | 1467 | 1279 |
| LMArena German | 1456 | 1234 |
| LMArena Japanese | 1405 | 1180 |
| LMArena Korean | 1416 | 1188 |
| LMArena Russian | 1457 | 1264 |
| LMArena Spanish | 1447 | 1255 |

## Instruction Following

- Grok 4.20 Multi-Agent: 75.0 (#80)
- Qwen2.5 72B Instruct: 65.7 (#216)

| Benchmark | Grok 4.20 Multi-Agent | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1420 | 1254 |
| IFEval | — | 80.6% |

## Long Context

- Grok 4.20 Multi-Agent: 43.7 (#84)
- Qwen2.5 72B Instruct: 38.9 (#183)

| Benchmark | Grok 4.20 Multi-Agent | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1430 | 1282 |

## Writing & Preference

- Grok 4.20 Multi-Agent: 64.1 (#57)
- Qwen2.5 72B Instruct: 46.9 (#210)

| Benchmark | Grok 4.20 Multi-Agent | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1450 | 1269 |
| LMArena Creative Writing | 1435 | 1221 |
| LMArena Multi-Turn | 1451 | 1272 |
| WildBench | — | 80.2% |

## FAQ

### Is Grok 4.20 Multi-Agent better than Qwen2.5 72B Instruct?

Grok 4.20 Multi-Agent is the stronger model overall, scoring 45.5 to 31.5 on the Noometry Index.

### Which is cheaper, Grok 4.20 Multi-Agent or Qwen2.5 72B Instruct?

Grok 4.20 Multi-Agent is cheaper. It lists at $1.25 per million input tokens and $2.50 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.

### Is Grok 4.20 Multi-Agent or Qwen2.5 72B Instruct better for coding?

Grok 4.20 Multi-Agent scores higher on coding benchmarks: 42.4 versus 32.8 in the Noometry coding category.

### Which has the bigger context window?

Grok 4.20 Multi-Agent does, with 1M tokens against 131K.

### How many benchmarks do Grok 4.20 Multi-Agent and Qwen2.5 72B Instruct share?

17 benchmarks have published results for both models. Grok 4.20 Multi-Agent has 20 scored results on Noometry and Qwen2.5 72B Instruct has 43.