# GPT-6 Luna vs Llama 4 Scout

> GPT-6 Luna is the stronger model overall, scoring 53.3 to 27.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-6-luna-vs-llama-4-scout
- Last updated: 2026-10-12
- Shared benchmarks: 27

## Summary

- They share 27 benchmarks with published results for both. GPT-6 Luna scores higher in 10 categories and Llama 4 Scout in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 19.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.9% for GPT-6 Luna and 7.8% for Llama 4 Scout.
- Llama 4 Scout is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.10 / $0.50 for GPT-6 Luna.
- GPT-6 Luna accepts more context: 1.05M tokens versus 128K.
- Llama 4 Scout has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-6 Luna | Llama 4 Scout |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 53.3 | 27.7 |
| Rank | 36 | 330 |
| Context | 1.05M | 128K |
| Input $/M | $0.10 | $0.10 |
| Output $/M | $0.50 | $0.30 |
| Weights | Proprietary | Open |

## Coding

- GPT-6 Luna: 55.5 (#25)
- Llama 4 Scout: 20.2 (#339)

| Benchmark | GPT-6 Luna | Llama 4 Scout |
|---|---|---|
| SciCode | 54.6% | 17% |
| LMArena Coding | 1439 | 1286 |
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| SWE-bench Verified (bash only) | — | 9.1% |
| LMArena WebDev | 1581 | — |
| BigCodeBench Complete | — | 43.1% |
| ALE-Bench | 1,577 | — |

## Agentic & Tool Use

- GPT-6 Luna: 33.3 (#54)
- Llama 4 Scout: 24.6 (#119)

| Benchmark | GPT-6 Luna | Llama 4 Scout |
|---|---|---|
| APEX-Agents | 44.3% | — |
| Berkeley Function Calling Leaderboard | — | 28.1% |
| GDP.pdf | 23% | — |

## Reasoning

- GPT-6 Luna: 48.2 (#41)
- Llama 4 Scout: 9.1 (#345)

| Benchmark | GPT-6 Luna | Llama 4 Scout |
|---|---|---|
| ARC-AGI-2 | 59.3% | 0% |
| ARC-AGI-1 | 86.7% | 0.5% |
| CritPt | 19.4% | 0% |
| LMArena Hard Prompts | 1411 | 1266 |
| DTBench | 90.1% | 57.9% |
| LMCA | 44.5% | 12% |
| Epoch Capabilities Index | 156.28 | 129.64 |
| Kagi LLM Benchmark | — | 36.9% |
| NYT Connections (extended) | 68.7% | — |
| Chess Puzzles | 31% | — |
| Mystery Game Puzzles | 7% | — |
| ForecastBench | — | 57.5 |

## Math

- GPT-6 Luna: 76.1 (#15)
- Llama 4 Scout: 19.6 (#286)

| Benchmark | GPT-6 Luna | Llama 4 Scout |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 7.8% |
| LMArena Math | 1416 | 1287 |
| FrontierMath (Tiers 1-3) | 78.9% | — |
| FrontierMath Tier 4 | 56.1% | — |
| ProofBench | 64% | — |
| Omni-MATH | — | 37.3% |
| MATH Level 5 | — | 62.3% |
| FrontierMath (Feb 2025 set) | — | 0% |

## Knowledge

- GPT-6 Luna: 57.0 (#41)
- Llama 4 Scout: 31.9 (#217)

| Benchmark | GPT-6 Luna | Llama 4 Scout |
|---|---|---|
| GPQA Diamond | 90.5% | 51.8% |
| LMArena Expert | 1444 | 1235 |
| SimpleQA Verified | 41.4% | — |
| MMLU-Pro | — | 74.2% |
| Vectara Hallucination Rate | — | 7.7% |
| GPQA (HELM) | — | 50.7% |

## Multimodal

- GPT-6 Luna: 42.4 (#30)
- Llama 4 Scout: 32.2 (#102)

| Benchmark | GPT-6 Luna | Llama 4 Scout |
|---|---|---|
| LMArena Vision | 1217 | 1118 |
| Blueprint-Bench 2 | 31.2% | — |
| Furniture Assembly | 44.2% | — |
| SpatialViz-Bench | — | 34.2% |

## Multilingual

- GPT-6 Luna: 50.5 (#117)
- Llama 4 Scout: 41.0 (#212)

| Benchmark | GPT-6 Luna | Llama 4 Scout |
|---|---|---|
| LMArena Non-English | 1386 | 1252 |
| LMArena Chinese | 1433 | 1255 |
| LMArena French | 1420 | 1282 |
| LMArena German | 1369 | 1272 |
| LMArena Japanese | 1369 | 1206 |
| LMArena Korean | 1360 | 1207 |
| LMArena Russian | 1394 | 1263 |
| LMArena Spanish | 1393 | 1278 |

## Instruction Following

- GPT-6 Luna: 74.3 (#99)
- Llama 4 Scout: 65.8 (#217)

| Benchmark | GPT-6 Luna | Llama 4 Scout |
|---|---|---|
| LMArena Instruction Following | 1409 | 1248 |
| IFEval | — | 81.8% |

## Long Context

- GPT-6 Luna: 43.0 (#111)
- Llama 4 Scout: 27.5 (#294)

| Benchmark | GPT-6 Luna | Llama 4 Scout |
|---|---|---|
| LMArena Longer Query | 1409 | 1265 |
| Fiction.LiveBench | — | 36% |

## Writing & Preference

- GPT-6 Luna: 58.3 (#119)
- Llama 4 Scout: 37.0 (#261)

| Benchmark | GPT-6 Luna | Llama 4 Scout |
|---|---|---|
| LMArena Text | 1391 | 1279 |
| LMArena Creative Writing | 1363 | 1249 |
| LMArena Multi-Turn | 1396 | 1280 |
| EQ-Bench Creative Writing | — | 783 |
| WildBench | — | 78% |

## FAQ

### Is GPT-6 Luna better than Llama 4 Scout?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 27.7 on the Noometry Index.

### Which is cheaper, GPT-6 Luna or Llama 4 Scout?

Llama 4 Scout is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; GPT-6 Luna lists at $0.10 and $0.50.

### Is GPT-6 Luna or Llama 4 Scout better for coding?

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 20.2 in the Noometry coding category.

### Which has the bigger context window?

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

### How many benchmarks do GPT-6 Luna and Llama 4 Scout share?

27 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Llama 4 Scout has 43.