# GPT-6 Luna vs Mistral 7B

> GPT-6 Luna is the stronger model overall, scoring 53.3 to 23.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-6-luna-vs-mistral-7b
- Last updated: 2026-10-12
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. GPT-6 Luna scores higher in 8 categories and Mistral 7B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 8.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.9% for GPT-6 Luna and 0.3% for Mistral 7B.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $0.25 / $0.25 for Mistral 7B.
- GPT-6 Luna accepts more context: 1.05M tokens versus 8K.
- Mistral 7B has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-6 Luna | Mistral 7B |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 53.3 | 23.0 |
| Rank | 36 | 351 |
| Context | 1.05M | 8K |
| Input $/M | $0.10 | $0.25 |
| Output $/M | $0.50 | $0.25 |
| Weights | Proprietary | Open |

## Coding

- GPT-6 Luna: 55.5 (#25)
- Mistral 7B: 26.4 (#326)

| Benchmark | GPT-6 Luna | Mistral 7B |
|---|---|---|
| LMArena Coding | 1439 | 1082 |
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| LMArena WebDev | 1581 | — |
| SciCode | 54.6% | — |
| BigCodeBench Instruct | — | 19.5% |
| BigCodeBench Complete | — | 27.3% |
| ALE-Bench | 1,577 | — |
| HumanEval+ | — | 36% |
| MBPP+ | — | 42.1% |

## Agentic & Tool Use

- GPT-6 Luna: 33.3 (#54)
- Mistral 7B: —

| Benchmark | GPT-6 Luna | Mistral 7B |
|---|---|---|
| APEX-Agents | 44.3% | — |
| GDP.pdf | 23% | — |

## Reasoning

- GPT-6 Luna: 48.2 (#41)
- Mistral 7B: 13.1 (#336)

| Benchmark | GPT-6 Luna | Mistral 7B |
|---|---|---|
| Chess Puzzles | 31% | 0% |
| LMArena Hard Prompts | 1411 | 1067 |
| DTBench | 90.1% | 42.5% |
| Epoch Capabilities Index | 156.28 | 112.21 |
| ARC-AGI-2 | 59.3% | — |
| NYT Connections (extended) | 68.7% | — |
| ARC-AGI-1 | 86.7% | — |
| CritPt | 19.4% | — |
| Mystery Game Puzzles | 7% | — |
| LMCA | 44.5% | — |
| Adversarial NLI | — | 47.1% |
| BIG-Bench Hard | — | 56.1% |
| HellaSwag | — | 81% |
| PIQA | — | 83% |
| WinoGrande | — | 75.3% |

## Math

- GPT-6 Luna: 76.1 (#15)
- Mistral 7B: 8.1 (#325)

| Benchmark | GPT-6 Luna | Mistral 7B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 0.3% |
| LMArena Math | 1416 | 1085 |
| FrontierMath (Tiers 1-3) | 78.9% | — |
| FrontierMath Tier 4 | 56.1% | — |
| ProofBench | 64% | — |
| MATH Level 5 | — | 3.7% |
| GSM8K | — | 54.4% |

## Knowledge

- GPT-6 Luna: 57.0 (#41)
- Mistral 7B: 7.4 (#311)

| Benchmark | GPT-6 Luna | Mistral 7B |
|---|---|---|
| GPQA Diamond | 90.5% | 15.2% |
| LMArena Expert | 1444 | 1036 |
| SimpleQA Verified | 41.4% | — |
| ARC (AI2) Challenge | — | 78.6% |
| BoolQ | — | 87.4% |
| MMLU | — | 62.5% |
| OpenBookQA | — | 79.8% |
| TriviaQA | — | 75.2% |

## Multimodal

- GPT-6 Luna: 42.4 (#30)
- Mistral 7B: —

| Benchmark | GPT-6 Luna | Mistral 7B |
|---|---|---|
| LMArena Vision | 1217 | — |
| Blueprint-Bench 2 | 31.2% | — |
| Furniture Assembly | 44.2% | — |

## Multilingual

- GPT-6 Luna: 50.5 (#117)
- Mistral 7B: 25.8 (#283)

| Benchmark | GPT-6 Luna | Mistral 7B |
|---|---|---|
| LMArena Non-English | 1386 | 1012 |
| LMArena Chinese | 1433 | 1009 |
| LMArena French | 1420 | 1037 |
| LMArena German | 1369 | 987 |
| LMArena Japanese | 1369 | 878 |
| LMArena Russian | 1394 | 1018 |
| LMArena Spanish | 1393 | 1026 |
| LMArena Korean | 1360 | — |

## Instruction Following

- GPT-6 Luna: 74.3 (#99)
- Mistral 7B: 54.2 (#280)

| Benchmark | GPT-6 Luna | Mistral 7B |
|---|---|---|
| LMArena Instruction Following | 1409 | 1060 |

## Long Context

- GPT-6 Luna: 43.0 (#111)
- Mistral 7B: 32.2 (#271)

| Benchmark | GPT-6 Luna | Mistral 7B |
|---|---|---|
| LMArena Longer Query | 1409 | 1060 |

## Writing & Preference

- GPT-6 Luna: 58.3 (#119)
- Mistral 7B: 30.7 (#286)

| Benchmark | GPT-6 Luna | Mistral 7B |
|---|---|---|
| LMArena Text | 1391 | 1090 |
| LMArena Creative Writing | 1363 | 1068 |
| LMArena Multi-Turn | 1396 | 1062 |

## FAQ

### Is GPT-6 Luna better than Mistral 7B?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 23.0 on the Noometry Index.

### Which is cheaper, GPT-6 Luna or Mistral 7B?

GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Mistral 7B lists at $0.25 and $0.25.

### Is GPT-6 Luna or Mistral 7B better for coding?

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 26.4 in the Noometry coding category.

### Which has the bigger context window?

GPT-6 Luna does, with 1.05M tokens against 8K.

### How many benchmarks do GPT-6 Luna and Mistral 7B share?

21 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Mistral 7B has 37.