# GPT-4 vs Mistral 7B

> GPT-4 is the stronger model overall, scoring 29.1 to 23.0 on the Noometry Index. Mistral 7B costs 150× less per token, which makes it the better buy when GPT-4's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-4-vs-mistral-7b
- Last updated: 2026-10-11
- Shared benchmarks: 31

## Summary

- They share 31 benchmarks with published results for both. GPT-4 scores higher in 8 categories and Mistral 7B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where GPT-4 leads 40.6 to 25.8.
- The biggest single-benchmark swing is BigCodeBench Complete: 57.2% for GPT-4 and 27.3% for Mistral 7B.
- Mistral 7B is cheaper at $0.25 / $0.25 per million input/output tokens, against $30 / $60 for GPT-4.
- GPT-4 accepts more context: 8K tokens versus 8K.
- Mistral 7B has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4 | Mistral 7B |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 29.1 | 23.0 |
| Rank | 316 | 351 |
| Context | 8K | 8K |
| Input $/M | $30 | $0.25 |
| Output $/M | $60 | $0.25 |
| Weights | Proprietary | Open |

## Coding

- GPT-4: 31.6 (#283)
- Mistral 7B: 26.4 (#326)

| Benchmark | GPT-4 | Mistral 7B |
|---|---|---|
| BigCodeBench Instruct | 46% | 19.5% |
| LMArena Coding | 1254 | 1082 |
| BigCodeBench Complete | 57.2% | 27.3% |
| HumanEval+ | 79.3% | 36% |
| WeirdML | 12.4% | — |
| MBPP+ | — | 42.1% |

## Agentic & Tool Use

- GPT-4: —
- Mistral 7B: —

| Benchmark | GPT-4 | Mistral 7B |
|---|---|---|
| METR Time Horizons | 36.1% | — |

## Reasoning

- GPT-4: 17.8 (#289)
- Mistral 7B: 13.1 (#336)

| Benchmark | GPT-4 | Mistral 7B |
|---|---|---|
| Chess Puzzles | 4% | 0% |
| LMArena Hard Prompts | 1241 | 1067 |
| DTBench | 62.7% | 42.5% |
| BIG-Bench Hard | 75.1% | 56.1% |
| Epoch Capabilities Index | 125.89 | 112.21 |
| HellaSwag | 95.3% | 81% |
| WinoGrande | 87.5% | 75.3% |
| Mystery Game Puzzles | 12% | — |
| LMCA | 17.1% | — |
| Adversarial NLI | — | 47.1% |
| ForecastBench | 57.8 | — |
| PIQA | — | 83% |

## Math

- GPT-4: 10.8 (#309)
- Mistral 7B: 8.1 (#325)

| Benchmark | GPT-4 | Mistral 7B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.1% | 0.3% |
| LMArena Math | 1269 | 1085 |
| MATH Level 5 | 23% | 3.7% |
| GSM8K | 92% | 54.4% |

## Knowledge

- GPT-4: 18.4 (#282)
- Mistral 7B: 7.4 (#311)

| Benchmark | GPT-4 | Mistral 7B |
|---|---|---|
| GPQA Diamond | 35.7% | 15.2% |
| LMArena Expert | 1211 | 1036 |
| MMLU | 86.4% | 62.5% |
| TriviaQA | 84.8% | 75.2% |
| ARC (AI2) Challenge | — | 78.6% |
| BoolQ | — | 87.4% |
| OpenBookQA | — | 79.8% |

## Multilingual

- GPT-4: 40.6 (#215)
- Mistral 7B: 25.8 (#283)

| Benchmark | GPT-4 | Mistral 7B |
|---|---|---|
| LMArena Non-English | 1246 | 1012 |
| LMArena Chinese | 1242 | 1009 |
| LMArena French | 1283 | 1037 |
| LMArena German | 1251 | 987 |
| LMArena Japanese | 1209 | 878 |
| LMArena Russian | 1251 | 1018 |
| LMArena Spanish | 1261 | 1026 |
| LMArena Korean | 1184 | — |

## Instruction Following

- GPT-4: 65.3 (#222)
- Mistral 7B: 54.2 (#280)

| Benchmark | GPT-4 | Mistral 7B |
|---|---|---|
| LMArena Instruction Following | 1241 | 1060 |

## Long Context

- GPT-4: 37.7 (#212)
- Mistral 7B: 32.2 (#271)

| Benchmark | GPT-4 | Mistral 7B |
|---|---|---|
| LMArena Longer Query | 1244 | 1060 |

## Writing & Preference

- GPT-4: 34.9 (#268)
- Mistral 7B: 30.7 (#286)

| Benchmark | GPT-4 | Mistral 7B |
|---|---|---|
| LMArena Text | 1263 | 1090 |
| LMArena Creative Writing | 1244 | 1068 |
| LMArena Multi-Turn | 1257 | 1062 |
| EQ-Bench Creative Writing | 752 | — |

## FAQ

### Is GPT-4 better than Mistral 7B?

GPT-4 is the stronger model overall, scoring 29.1 to 23.0 on the Noometry Index. Mistral 7B costs 150× less per token, which makes it the better buy when GPT-4's lead doesn't matter for your workload.

### Which is cheaper, GPT-4 or Mistral 7B?

Mistral 7B is cheaper. It lists at $0.25 per million input tokens and $0.25 per million output tokens; GPT-4 lists at $30 and $60.

### Is GPT-4 or Mistral 7B better for coding?

GPT-4 scores higher on coding benchmarks: 31.6 versus 26.4 in the Noometry coding category.

### Which has the bigger context window?

GPT-4 does, with 8K tokens against 8K.

### How many benchmarks do GPT-4 and Mistral 7B share?

31 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Mistral 7B has 37.