# GPT-6 Sol vs Llama 4 Maverick

> GPT-6 Sol is the stronger model overall, scoring 49.2 to 30.6 on the Noometry Index. Llama 4 Maverick costs 13× less per token, which makes it the better buy when GPT-6 Sol's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-6-sol-vs-llama-4-maverick
- Last updated: 2026-10-10
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. GPT-6 Sol scores higher in 10 categories and Llama 4 Maverick in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Sol leads 49.3 to 10.1.
- The biggest single-benchmark swing is NYT Connections (extended): 90.1% for GPT-6 Sol and 8% for Llama 4 Maverick.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $2 / $10 for GPT-6 Sol.
- GPT-6 Sol accepts more context: 1.05M tokens versus 128K.
- Llama 4 Maverick has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-6 Sol | Llama 4 Maverick |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 49.2 | 30.6 |
| Rank | 46 | 276 |
| Context | 1.05M | 128K |
| Input $/M | $2 | $0.19 |
| Output $/M | $10 | $0.65 |
| Weights | Proprietary | Open |

## Coding

- GPT-6 Sol: 59.6 (#11)
- Llama 4 Maverick: 26.3 (#318)

| Benchmark | GPT-6 Sol | Llama 4 Maverick |
|---|---|---|
| LMArena Coding | 1453 | 1302 |
| DeepSWE | 68.8% | — |
| FrontierCode | 49.3% | — |
| SWE-bench Verified (bash only) | — | 21% |
| Aider Polyglot | — | 15.6% |
| LMArena WebDev | 1692 | — |
| SciCode | — | 33.1% |
| WeirdML | — | 24.5% |
| BigCodeBench Instruct | — | 49.7% |
| BigCodeBench Complete | — | 61.4% |
| ALE-Bench | — | 172.97 |

## Agentic & Tool Use

- GPT-6 Sol: 33.5 (#49)
- Llama 4 Maverick: 27.9 (#86)

| Benchmark | GPT-6 Sol | Llama 4 Maverick |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.3% |
| GDP.pdf | 28% | — |

## Reasoning

- GPT-6 Sol: 49.3 (#34)
- Llama 4 Maverick: 10.1 (#337)

| Benchmark | GPT-6 Sol | Llama 4 Maverick |
|---|---|---|
| NYT Connections (extended) | 90.1% | 8% |
| LMArena Hard Prompts | 1420 | 1281 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 27.7% |
| Kagi LLM Benchmark | — | 55.9% |
| ARC-AGI-1 | — | 4.4% |
| CritPt | — | 0% |
| EnigmaEval | — | 0.6% |
| EBR-Bench | 53.3% | — |
| DTBench | — | 61.9% |
| LMCA | — | 15.9% |
| Epoch Capabilities Index | — | 132.2 |
| ForecastBench | — | 57.5 |

## Math

- GPT-6 Sol: 35.8 (#135)
- Llama 4 Maverick: 24.6 (#256)

| Benchmark | GPT-6 Sol | Llama 4 Maverick |
|---|---|---|
| LMArena Math | 1404 | 1299 |
| OTIS Mock AIME 2024-2025 | — | 20.6% |
| Omni-MATH | — | 42.2% |
| MATH Level 5 | — | 73% |
| FrontierMath (Feb 2025 set) | — | 0.7% |

## Knowledge

- GPT-6 Sol: 43.0 (#87)
- Llama 4 Maverick: 33.3 (#196)

| Benchmark | GPT-6 Sol | Llama 4 Maverick |
|---|---|---|
| Vectara Hallucination Rate | 6.5% | 8.2% |
| LMArena Expert | 1449 | 1259 |
| GPQA Diamond | — | 67% |
| Humanity's Last Exam | — | 5.7% |
| MMLU-Pro | — | 81% |
| Confabulations | — | 22.6% |
| GPQA (HELM) | — | 65% |

## Multimodal

- GPT-6 Sol: 47.6 (#9)
- Llama 4 Maverick: 30.9 (#100)

| Benchmark | GPT-6 Sol | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | — | 1141 |
| GeoBench | — | 52% |
| Furniture Assembly | 58.3% | — |
| SpatialViz-Bench | — | 31.8% |

## Multilingual

- GPT-6 Sol: 51.0 (#113)
- Llama 4 Maverick: 42.4 (#190)

| Benchmark | GPT-6 Sol | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1389 | 1269 |
| LMArena Chinese | 1416 | 1277 |
| LMArena Russian | 1407 | 1286 |
| LMArena Spanish | 1377 | 1293 |
| LMArena French | — | 1258 |
| LMArena German | — | 1291 |
| LMArena Japanese | — | 1207 |
| LMArena Korean | — | 1203 |

## Instruction Following

- GPT-6 Sol: 74.4 (#97)
- Llama 4 Maverick: 71.9 (#140)

| Benchmark | GPT-6 Sol | Llama 4 Maverick |
|---|---|---|
| LMArena Instruction Following | 1406 | 1267 |
| IFEval | — | 90.8% |

## Long Context

- GPT-6 Sol: 43.2 (#100)
- Llama 4 Maverick: 31.4 (#274)

| Benchmark | GPT-6 Sol | Llama 4 Maverick |
|---|---|---|
| LMArena Longer Query | 1415 | 1280 |
| Fiction.LiveBench | — | 46.2% |

## Writing & Preference

- GPT-6 Sol: 72.3 (#15)
- Llama 4 Maverick: 39.0 (#247)

| Benchmark | GPT-6 Sol | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1396 | 1287 |
| LMArena Creative Writing | 1392 | 1266 |
| EQ-Bench Creative Writing | 2125 | 860 |
| LMArena Multi-Turn | 1405 | 1289 |
| Short-Story Creative Writing | — | 62% |
| WildBench | — | 80% |

## FAQ

### Is GPT-6 Sol better than Llama 4 Maverick?

GPT-6 Sol is the stronger model overall, scoring 49.2 to 30.6 on the Noometry Index. Llama 4 Maverick costs 13× less per token, which makes it the better buy when GPT-6 Sol's lead doesn't matter for your workload.

### Which is cheaper, GPT-6 Sol or Llama 4 Maverick?

Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; GPT-6 Sol lists at $2 and $10.

### Is GPT-6 Sol or Llama 4 Maverick better for coding?

GPT-6 Sol scores higher on coding benchmarks: 59.6 versus 26.3 in the Noometry coding category.

### Which has the bigger context window?

GPT-6 Sol does, with 1.05M tokens against 128K.

### How many benchmarks do GPT-6 Sol and Llama 4 Maverick share?

16 benchmarks have published results for both models. GPT-6 Sol has 22 scored results on Noometry and Llama 4 Maverick has 54.