# Ministral 3B vs Qwen3-4B

> Qwen3-4B is the stronger model overall, scoring 31.9 to 26.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/ministral-3b-vs-qwen3-4b
- Last updated: 2026-10-12
- Shared benchmarks: 2

## Summary

- They share 2 benchmarks with published results for both. Ministral 3B scores higher in 0 categories and Qwen3-4B in 3 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3-4B leads 33.0 to 10.4.
- The biggest single-benchmark swing is GPQA Diamond: 25.3% for Ministral 3B and 52.3% for Qwen3-4B.

## Snapshot

| | Ministral 3B | Qwen3-4B |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 26.2 | 31.9 |
| Rank | 338 | 264 |
| Context | 131K | — |
| Input $/M | $0.10 | — |
| Output $/M | $0.10 | — |
| Weights | Open | Open |

## Agentic & Tool Use

- Ministral 3B: —
- Qwen3-4B: 27.6 (#100)

| Benchmark | Ministral 3B | Qwen3-4B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 35.7% |

## Reasoning

- Ministral 3B: 18.4 (#282)
- Qwen3-4B: 19.2 (#268)

| Benchmark | Ministral 3B | Qwen3-4B |
|---|---|---|
| Chess Puzzles | — | 4% |
| DTBench | 51.7% | — |
| LMCA | 5.5% | — |
| Epoch Capabilities Index | 118.1 | — |

## Math

- Ministral 3B: 26.6 (#258)
- Qwen3-4B: 29.7 (#240)

| Benchmark | Ministral 3B | Qwen3-4B |
|---|---|---|
| MathArena Final-Answer Competitions | — | 38.5% |
| OTIS Mock AIME 2024-2025 | — | 52.2% |
| MATH Level 5 | 14.4% | — |

## Knowledge

- Ministral 3B: 10.4 (#302)
- Qwen3-4B: 33.0 (#208)

| Benchmark | Ministral 3B | Qwen3-4B |
|---|---|---|
| GPQA Diamond | 25.3% | 52.3% |
| Vectara Hallucination Rate | 7.3% | 5.7% |

## FAQ

### Is Ministral 3B better than Qwen3-4B?

Qwen3-4B is the stronger model overall, scoring 31.9 to 26.2 on the Noometry Index.

### How many benchmarks do Ministral 3B and Qwen3-4B share?

2 benchmarks have published results for both models. Ministral 3B has 6 scored results on Noometry and Qwen3-4B has 6.