# GPT-4o mini vs Llama 4 Maverick

> Llama 4 Maverick is the stronger model overall, scoring 30.6 to 25.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4o-mini-vs-llama-4-maverick
- Last updated: 2026-10-10
- Shared benchmarks: 40

## Summary

- They share 40 benchmarks with published results for both. GPT-4o mini scores higher in 2 categories and Llama 4 Maverick in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama 4 Maverick leads 33.3 to 17.7.
- The biggest single-benchmark swing is GPQA Diamond: 37.7% for GPT-4o mini and 67% for Llama 4 Maverick.
- GPT-4o mini is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.19 / $0.65 for Llama 4 Maverick.
- Llama 4 Maverick has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4o mini | Llama 4 Maverick |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 25.4 | 30.6 |
| Rank | 337 | 276 |
| Context | 128K | 128K |
| Input $/M | $0.15 | $0.19 |
| Output $/M | $0.60 | $0.65 |
| Weights | Proprietary | Open |

## Coding

- GPT-4o mini: 21.6 (#330)
- Llama 4 Maverick: 26.3 (#318)

| Benchmark | GPT-4o mini | Llama 4 Maverick |
|---|---|---|
| Aider Polyglot | 3.6% | 15.6% |
| WeirdML | 11.8% | 24.5% |
| BigCodeBench Instruct | 46.1% | 49.7% |
| LMArena Coding | 1290 | 1302 |
| BigCodeBench Complete | 57.4% | 61.4% |
| SWE-bench Verified (bash only) | — | 21% |
| SciCode | — | 33.1% |
| LiveBench Coding | 43.1% | — |
| ALE-Bench | — | 172.97 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 72.2% | — |

## Agentic & Tool Use

- GPT-4o mini: 27.2 (#95)
- Llama 4 Maverick: 27.9 (#86)

| Benchmark | GPT-4o mini | Llama 4 Maverick |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.3% |
| BALROG | 17.4% | — |

## Reasoning

- GPT-4o mini: 8.7 (#343)
- Llama 4 Maverick: 10.1 (#337)

| Benchmark | GPT-4o mini | Llama 4 Maverick |
|---|---|---|
| ARC-AGI-2 | 0% | 0% |
| SimpleBench | 10.7% | 27.7% |
| Kagi LLM Benchmark | 28.8% | 55.9% |
| LMArena Hard Prompts | 1267 | 1281 |
| DTBench | 54.4% | 61.9% |
| LMCA | 10.4% | 15.9% |
| Epoch Capabilities Index | 126.56 | 132.2 |
| NYT Connections (extended) | — | 8% |
| ARC-AGI-1 | — | 4.4% |
| CritPt | — | 0% |
| Chess Puzzles | 0% | — |
| EnigmaEval | — | 0.6% |
| LiveBench Reasoning | 32.8% | — |
| Mystery Game Puzzles | 12% | — |
| LiveBench Data Analysis | 50% | — |
| ForecastBench | — | 57.5 |
| LiveBench | 41.3% | — |
| PIQA | 88.7% | — |

## Math

- GPT-4o mini: 9.9 (#307)
- Llama 4 Maverick: 24.6 (#256)

| Benchmark | GPT-4o mini | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.9% | 20.6% |
| Omni-MATH | 28% | 42.2% |
| LMArena Math | 1267 | 1299 |
| MATH Level 5 | 52.6% | 73% |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| LiveBench Math | 36.3% | — |
| FrontierMath (Feb 2025 set) | — | 0.7% |
| GSM8K | 91.3% | — |

## Knowledge

- GPT-4o mini: 17.7 (#279)
- Llama 4 Maverick: 33.3 (#196)

| Benchmark | GPT-4o mini | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 37.7% | 67% |
| MMLU-Pro | 60.3% | 81% |
| Confabulations | 37.2% | 22.6% |
| GPQA (HELM) | 36.8% | 65% |
| LMArena Expert | 1235 | 1259 |
| Humanity's Last Exam | — | 5.7% |
| SimpleQA Verified | 8.3% | — |
| Vectara Hallucination Rate | — | 8.2% |
| BoolQ | 88.7% | — |
| MMLU | 81.8% | — |

## Multimodal

- GPT-4o mini: 25.3 (#117)
- Llama 4 Maverick: 30.9 (#100)

| Benchmark | GPT-4o mini | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | 1066 | 1141 |
| GeoBench | 64% | 52% |
| Video-MME | 64.8% | — |
| VPCT | 34% | — |
| SpatialViz-Bench | — | 31.8% |

## Multilingual

- GPT-4o mini: 42.2 (#194)
- Llama 4 Maverick: 42.4 (#190)

| Benchmark | GPT-4o mini | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1266 | 1269 |
| LMArena Chinese | 1264 | 1277 |
| LMArena French | 1296 | 1258 |
| LMArena German | 1272 | 1291 |
| LMArena Japanese | 1216 | 1207 |
| LMArena Korean | 1195 | 1203 |
| LMArena Russian | 1275 | 1286 |
| LMArena Spanish | 1276 | 1293 |

## Instruction Following

- GPT-4o mini: 62.1 (#234)
- Llama 4 Maverick: 71.9 (#140)

| Benchmark | GPT-4o mini | Llama 4 Maverick |
|---|---|---|
| IFEval | 78.2% | 90.8% |
| LMArena Instruction Following | 1259 | 1267 |
| LiveBench Instruction Following | 56.8% | — |

## Long Context

- GPT-4o mini: 39.1 (#181)
- Llama 4 Maverick: 31.4 (#274)

| Benchmark | GPT-4o mini | Llama 4 Maverick |
|---|---|---|
| LMArena Longer Query | 1289 | 1280 |
| Fiction.LiveBench | — | 46.2% |

## Writing & Preference

- GPT-4o mini: 39.7 (#243)
- Llama 4 Maverick: 39.0 (#247)

| Benchmark | GPT-4o mini | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1286 | 1287 |
| LMArena Creative Writing | 1268 | 1266 |
| Short-Story Creative Writing | 67.2% | 62% |
| EQ-Bench Creative Writing | 873 | 860 |
| WildBench | 79.1% | 80% |
| LMArena Multi-Turn | 1285 | 1289 |
| LiveBench Language | 28.6% | — |

## FAQ

### Is GPT-4o mini better than Llama 4 Maverick?

Llama 4 Maverick is the stronger model overall, scoring 30.6 to 25.4 on the Noometry Index.

### Which is cheaper, GPT-4o mini or Llama 4 Maverick?

GPT-4o mini is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Llama 4 Maverick lists at $0.19 and $0.65.

### Is GPT-4o mini or Llama 4 Maverick better for coding?

Llama 4 Maverick scores higher on coding benchmarks: 26.3 versus 21.6 in the Noometry coding category.

### Which has the bigger context window?

Both accept 128K tokens.

### How many benchmarks do GPT-4o mini and Llama 4 Maverick share?

40 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and Llama 4 Maverick has 54.