# GPT-4.1 mini vs Grok 4.20 (Non-Reasoning)

> Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 33.5 on the Noometry Index. GPT-4.1 mini costs 2.2× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-4-1-mini-vs-grok-4-20
- Last updated: 2026-10-10
- Shared benchmarks: 31

## Summary

- They share 31 benchmarks with published results for both. GPT-4.1 mini scores higher in 1 category and Grok 4.20 (Non-Reasoning) in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.2 to 10.9.
- The biggest single-benchmark swing is ARC-AGI-1: 3.5% for GPT-4.1 mini and 89.5% for Grok 4.20 (Non-Reasoning).
- GPT-4.1 mini is cheaper at $0.40 / $1.60 per million input/output tokens, against $1.25 / $2.50 for Grok 4.20 (Non-Reasoning).
- GPT-4.1 mini accepts more context: 1.05M tokens versus 1M.

## Snapshot

| | GPT-4.1 mini | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 33.5 | 48.5 |
| Rank | 232 | 48 |
| Context | 1.05M | 1M |
| Input $/M | $0.40 | $1.25 |
| Output $/M | $1.60 | $2.50 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4.1 mini: 30.4 (#286)
- Grok 4.20 (Non-Reasoning): 41.8 (#104)

| Benchmark | GPT-4.1 mini | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| WeirdML | 37.6% | 52.3% |
| LMArena Coding | 1367 | 1458 |
| SWE-bench Verified (bash only) | 23.9% | — |
| Aider Polyglot | 32.4% | — |
| LMArena WebDev | — | 1374 |
| SciCode | 40.4% | — |
| BigCodeBench Instruct | 48.9% | — |
| CadEval | 16% | — |
| ALE-Bench | — | 1,150 |

## Agentic & Tool Use

- GPT-4.1 mini: 33.0 (#52)
- Grok 4.20 (Non-Reasoning): 34.3 (#43)

| Benchmark | GPT-4.1 mini | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Terminal-Bench | — | 57.3% |
| Berkeley Function Calling Leaderboard | 50.5% | — |
| τ²-bench Banking | — | 18% |
| LMArena Search | — | 1189 |
| Vending-Bench 2 | — | 4,663 |

## Reasoning

- GPT-4.1 mini: 10.9 (#335)
- Grok 4.20 (Non-Reasoning): 52.2 (#30)

| Benchmark | GPT-4.1 mini | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| ARC-AGI-2 | 0% | 65.1% |
| Kagi LLM Benchmark | 48.6% | 75% |
| ARC-AGI-1 | 3.5% | 89.5% |
| Chess Puzzles | 7% | 24% |
| LMArena Hard Prompts | 1349 | 1451 |
| DTBench | 68.8% | 90.1% |
| LMCA | 21.1% | 38.7% |
| Epoch Capabilities Index | 135.01 | 152.01 |
| NYT Connections (extended) | — | 85.4% |
| CritPt | 0% | — |
| Thematic Generalization | — | 63.8% |
| Mystery Game Puzzles | 7% | — |
| ForecastBench | — | 61.4 |

## Math

- GPT-4.1 mini: 23.1 (#264)
- Grok 4.20 (Non-Reasoning): 48.0 (#54)

| Benchmark | GPT-4.1 mini | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6.7% | 44.9% |
| OTIS Mock AIME 2024-2025 | 44.7% | 92.2% |
| LMArena Math | 1343 | 1455 |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 14% |
| Omni-MATH | 49.1% | — |
| MATH Level 5 | 87.3% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |

## Knowledge

- GPT-4.1 mini: 34.7 (#183)
- Grok 4.20 (Non-Reasoning): 52.8 (#54)

| Benchmark | GPT-4.1 mini | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| GPQA Diamond | 65.8% | 89.3% |
| SimpleQA Verified | 12.7% | 30.2% |
| LMArena Expert | 1338 | 1440 |
| MMLU-Pro | 78.3% | — |
| GPQA (HELM) | 61.4% | — |

## Multimodal

- GPT-4.1 mini: 35.2 (#77)
- Grok 4.20 (Non-Reasoning): 33.2 (#88)

| Benchmark | GPT-4.1 mini | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Vision | 1181 | 1263 |
| Blueprint-Bench 2 | — | 0% |
| LMArena Document | — | 1416 |

## Multilingual

- GPT-4.1 mini: 45.9 (#161)
- Grok 4.20 (Non-Reasoning): 54.7 (#38)

| Benchmark | GPT-4.1 mini | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Non-English | 1318 | 1440 |
| LMArena Chinese | 1329 | 1481 |
| LMArena French | 1357 | 1477 |
| LMArena German | 1349 | 1463 |
| LMArena Japanese | 1291 | 1450 |
| LMArena Korean | 1298 | 1417 |
| LMArena Russian | 1324 | 1457 |
| LMArena Spanish | 1319 | 1442 |

## Instruction Following

- GPT-4.1 mini: 73.9 (#113)
- Grok 4.20 (Non-Reasoning): 75.0 (#79)

| Benchmark | GPT-4.1 mini | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Instruction Following | 1332 | 1420 |
| IFEval | 90.4% | — |

## Long Context

- GPT-4.1 mini: 31.8 (#270)
- Grok 4.20 (Non-Reasoning): 45.4 (#33)

| Benchmark | GPT-4.1 mini | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Longer Query | 1344 | 1436 |
| Fiction.LiveBench | 44.4% | — |
| CL-bench | — | 22.2% |
| CL-bench Life | — | 11.9% |

## Writing & Preference

- GPT-4.1 mini: 48.8 (#193)
- Grok 4.20 (Non-Reasoning): 65.9 (#42)

| Benchmark | GPT-4.1 mini | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Text | 1341 | 1451 |
| LMArena Creative Writing | 1300 | 1439 |
| EQ-Bench Creative Writing | 1147 | 1574 |
| LMArena Multi-Turn | 1354 | 1456 |
| WildBench | 83.8% | — |

## FAQ

### Is GPT-4.1 mini better than Grok 4.20 (Non-Reasoning)?

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 33.5 on the Noometry Index. GPT-4.1 mini costs 2.2× less per token, which makes it the better buy when Grok 4.20 (Non-Reasoning)'s lead doesn't matter for your workload.

### Which is cheaper, GPT-4.1 mini or Grok 4.20 (Non-Reasoning)?

GPT-4.1 mini is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; Grok 4.20 (Non-Reasoning) lists at $1.25 and $2.50.

### Is GPT-4.1 mini or Grok 4.20 (Non-Reasoning) better for coding?

Grok 4.20 (Non-Reasoning) scores higher on coding benchmarks: 41.8 versus 30.4 in the Noometry coding category.

### Which has the bigger context window?

GPT-4.1 mini does, with 1.05M tokens against 1M.

### How many benchmarks do GPT-4.1 mini and Grok 4.20 (Non-Reasoning) share?

31 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.