# GPT-4 vs Ministral 8B

> GPT-4 is the stronger model overall, scoring 28.9 to 27.8 on the Noometry Index. Ministral 8B costs 375× 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-ministral-8b
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. GPT-4 scores higher in 4 categories and Ministral 8B in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Ministral 8B leads 23.7 to 10.1.
- The biggest single-benchmark swing is DTBench: 62.7% for GPT-4 and 45.7% for Ministral 8B.
- Ministral 8B is cheaper at $0.10 / $0.10 per million input/output tokens, against $30 / $60 for GPT-4.
- Ministral 8B accepts more context: 128K tokens versus 8K.
- Ministral 8B has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4 | Ministral 8B |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 28.9 | 27.8 |
| Rank | 308 | 321 |
| Context | 8K | 128K |
| Input $/M | $30 | $0.10 |
| Output $/M | $60 | $0.10 |
| Weights | Proprietary | Open |

## Coding

- GPT-4: 31.2 (#277)
- Ministral 8B: 34.4 (#226)

| Benchmark | GPT-4 | Ministral 8B |
|---|---|---|
| LMArena Coding | 1254 | 1202 |
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |

## Agentic & Tool Use

- GPT-4: —
- Ministral 8B: 16.2 (#140)

| Benchmark | GPT-4 | Ministral 8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 11.1% |
| METR Time Horizons | 36.1% | — |

## Reasoning

- GPT-4: 17.3 (#285)
- Ministral 8B: 17.7 (#279)

| Benchmark | GPT-4 | Ministral 8B |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1191 |
| DTBench | 62.7% | 45.7% |
| Chess Puzzles | 4% | — |
| Mystery Game Puzzles | 12% | — |
| LMCA | 17.1% | — |
| BIG-Bench Hard | 75.1% | — |
| Epoch Capabilities Index | 125.88 | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |

## Math

- GPT-4: 10.1 (#304)
- Ministral 8B: 23.7 (#262)

| Benchmark | GPT-4 | Ministral 8B |
|---|---|---|
| LMArena Math | 1269 | 1188 |
| MATH Level 5 | 23% | 14.9% |
| OTIS Mock AIME 2024-2025 | 1.1% | — |
| GSM8K | 92% | — |

## Knowledge

- GPT-4: 18.4 (#277)
- Ministral 8B: 12.5 (#292)

| Benchmark | GPT-4 | Ministral 8B |
|---|---|---|
| GPQA Diamond | 35.7% | 27.1% |
| LMArena Expert | 1211 | 1170 |
| Vectara Hallucination Rate | — | 7.4% |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |

## Multilingual

- GPT-4: 40.8 (#210)
- Ministral 8B: 35.3 (#242)

| Benchmark | GPT-4 | Ministral 8B |
|---|---|---|
| LMArena Non-English | 1246 | 1165 |
| LMArena Chinese | 1242 | 1193 |
| LMArena Russian | 1251 | 1195 |
| LMArena French | 1282 | — |
| LMArena German | 1251 | — |
| LMArena Japanese | 1209 | — |
| LMArena Korean | 1184 | — |
| LMArena Spanish | 1261 | — |

## Instruction Following

- GPT-4: 65.5 (#217)
- Ministral 8B: 60.7 (#245)

| Benchmark | GPT-4 | Ministral 8B |
|---|---|---|
| LMArena Instruction Following | 1241 | 1161 |

## Long Context

- GPT-4: 37.7 (#207)
- Ministral 8B: 36.7 (#222)

| Benchmark | GPT-4 | Ministral 8B |
|---|---|---|
| LMArena Longer Query | 1244 | 1212 |

## Writing & Preference

- GPT-4: 35.0 (#263)
- Ministral 8B: 39.8 (#241)

| Benchmark | GPT-4 | Ministral 8B |
|---|---|---|
| LMArena Text | 1263 | 1191 |
| LMArena Creative Writing | 1244 | 1175 |
| LMArena Multi-Turn | 1257 | 1166 |
| EQ-Bench Creative Writing | 752 | — |

## FAQ

### Is GPT-4 better than Ministral 8B?

GPT-4 is the stronger model overall, scoring 28.9 to 27.8 on the Noometry Index. Ministral 8B costs 375× 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 Ministral 8B?

Ministral 8B is cheaper. It lists at $0.10 per million input tokens and $0.10 per million output tokens; GPT-4 lists at $30 and $60.

### Is GPT-4 or Ministral 8B better for coding?

Ministral 8B scores higher on coding benchmarks: 34.4 versus 31.2 in the Noometry coding category.

### Which has the bigger context window?

Ministral 8B does, with 128K tokens against 8K.

### How many benchmarks do GPT-4 and Ministral 8B share?

15 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Ministral 8B has 17.