# Grok-3 mini vs Grok 4.20 Multi-Agent

> Grok 4.20 Multi-Agent is the stronger model overall, scoring 45.5 to 41.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-3-mini-vs-grok-4-20-multi-agent
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Grok-3 mini scores higher in 3 categories and Grok 4.20 Multi-Agent in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 Multi-Agent leads 42.4 to 13.8.

## Snapshot

| | Grok-3 mini | Grok 4.20 Multi-Agent |
|---|---|---|
| Provider | xAI | xAI |
| Noometry Index | 41.0 | 45.5 |
| Rank | 130 | 62 |
| Context | — | 1M |
| Input $/M | — | $1.25 |
| Output $/M | — | $2.50 |
| Weights | Proprietary | Proprietary |

## Coding

- Grok-3 mini: 40.4 (#126)
- Grok 4.20 Multi-Agent: 42.4 (#91)

| Benchmark | Grok-3 mini | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Coding | 1380 | 1458 |
| Aider Polyglot | 49.3% | — |
| WeirdML | 42.6% | — |

## Agentic & Tool Use

- Grok-3 mini: —
- Grok 4.20 Multi-Agent: —

| Benchmark | Grok-3 mini | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Search | — | 1204 |

## Reasoning

- Grok-3 mini: 13.8 (#323)
- Grok 4.20 Multi-Agent: 42.4 (#44)

| Benchmark | Grok-3 mini | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Hard Prompts | 1374 | 1448 |
| ARC-AGI-2 | 0.4% | — |
| Kagi LLM Benchmark | 61.3% | — |
| NYT Connections (extended) | — | 89.6% |
| ARC-AGI-1 | 16.5% | — |
| Epoch Capabilities Index | 140.35 | — |

## Math

- Grok-3 mini: 40.3 (#78)
- Grok 4.20 Multi-Agent: 36.6 (#106)

| Benchmark | Grok-3 mini | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Math | 1386 | 1441 |
| OTIS Mock AIME 2024-2025 | 77.8% | — |
| Omni-MATH | 31.8% | — |
| MATH Level 5 | 90.9% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |

## Knowledge

- Grok-3 mini: 46.3 (#75)
- Grok 4.20 Multi-Agent: 39.7 (#116)

| Benchmark | Grok-3 mini | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Expert | 1395 | 1445 |
| GPQA Diamond | 76.3% | — |
| MMLU-Pro | 79.9% | — |
| Confabulations | 10.8% | — |
| GPQA (HELM) | 67.5% | — |

## Multimodal

- Grok-3 mini: —
- Grok 4.20 Multi-Agent: 39.8 (#44)

| Benchmark | Grok-3 mini | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Vision | — | 1259 |

## Multilingual

- Grok-3 mini: 48.3 (#141)
- Grok 4.20 Multi-Agent: 54.6 (#40)

| Benchmark | Grok-3 mini | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Non-English | 1352 | 1440 |
| LMArena Chinese | 1387 | 1475 |
| LMArena French | 1357 | 1467 |
| LMArena German | 1350 | 1456 |
| LMArena Japanese | 1343 | 1405 |
| LMArena Korean | 1336 | 1416 |
| LMArena Russian | 1352 | 1457 |
| LMArena Spanish | 1379 | 1447 |

## Instruction Following

- Grok-3 mini: 78.7 (#8)
- Grok 4.20 Multi-Agent: 75.0 (#80)

| Benchmark | Grok-3 mini | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Instruction Following | 1356 | 1420 |
| IFEval | 95.1% | — |

## Long Context

- Grok-3 mini: 41.0 (#142)
- Grok 4.20 Multi-Agent: 43.7 (#84)

| Benchmark | Grok-3 mini | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Longer Query | 1372 | 1430 |
| Fiction.LiveBench | 66.7% | — |

## Writing & Preference

- Grok-3 mini: 52.7 (#163)
- Grok 4.20 Multi-Agent: 64.1 (#57)

| Benchmark | Grok-3 mini | Grok 4.20 Multi-Agent |
|---|---|---|
| LMArena Text | 1370 | 1450 |
| LMArena Creative Writing | 1342 | 1435 |
| LMArena Multi-Turn | 1355 | 1451 |
| Short-Story Creative Writing | 73.5% | — |
| WildBench | 65.1% | — |

## FAQ

### Is Grok-3 mini better than Grok 4.20 Multi-Agent?

Grok 4.20 Multi-Agent is the stronger model overall, scoring 45.5 to 41.0 on the Noometry Index.

### Is Grok-3 mini or Grok 4.20 Multi-Agent better for coding?

Grok 4.20 Multi-Agent scores higher on coding benchmarks: 42.4 versus 40.4 in the Noometry coding category.

### How many benchmarks do Grok-3 mini and Grok 4.20 Multi-Agent share?

17 benchmarks have published results for both models. Grok-3 mini has 35 scored results on Noometry and Grok 4.20 Multi-Agent has 20.