# Mercury 2 vs MiniMax-M3

> MiniMax-M3 is the stronger model overall, scoring 43.8 to 39.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/mercury-2-vs-minimax-m3
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. Mercury 2 scores higher in 0 categories and MiniMax-M3 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 36.2.
- The biggest single-benchmark swing is SciCode: 38.7% for Mercury 2 and 47.1% for MiniMax-M3.
- Mercury 2 is cheaper at $0.25 / $0.75 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.
- MiniMax-M3 accepts more context: 1M tokens versus 128K.
- MiniMax-M3 has downloadable open weights; the other is API-only.

## Snapshot

| | Mercury 2 | MiniMax-M3 |
|---|---|---|
| Provider | Inception | MiniMax |
| Noometry Index | 39.1 | 43.8 |
| Rank | 175 | 85 |
| Context | 128K | 1M |
| Input $/M | $0.25 | $0.30 |
| Output $/M | $0.75 | $1.20 |
| Weights | Proprietary | Open |

## Coding

- Mercury 2: 33.5 (#255)
- MiniMax-M3: 41.8 (#118)

| Benchmark | Mercury 2 | MiniMax-M3 |
|---|---|---|
| LMArena WebDev | 1171 | 1482 |
| SciCode | 38.7% | 47.1% |
| LMArena Coding | 1391 | 1469 |
| ALE-Bench | 785.58 | 640.02 |
| FrontierCode | — | 14.7% |
| WeirdML | 43.2% | — |

## Agentic & Tool Use

- Mercury 2: —
- MiniMax-M3: 22.6 (#130)

| Benchmark | Mercury 2 | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |

## Reasoning

- Mercury 2: 23.8 (#170)
- MiniMax-M3: 30.1 (#87)

| Benchmark | Mercury 2 | MiniMax-M3 |
|---|---|---|
| CritPt | 0.8% | 3.7% |
| LMArena Hard Prompts | 1362 | 1447 |
| SimpleBench | — | 45.8% |
| NYT Connections (extended) | — | 65.1% |
| Chess Puzzles | — | 14% |
| Mystery Game Puzzles | — | 8% |
| DTBench | — | 78.9% |
| LMCA | — | 33.7% |
| Surface Evolver Bench | — | 55% |
| Epoch Capabilities Index | — | 146.95 |
| ForecastBench | — | 61.4 |

## Math

- Mercury 2: —
- MiniMax-M3: 40.0 (#95)

| Benchmark | Mercury 2 | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 71.1% |
| ProofBench | — | 18% |
| LMArena Math | — | 1429 |

## Knowledge

- Mercury 2: 36.2 (#172)
- MiniMax-M3: 58.4 (#35)

| Benchmark | Mercury 2 | MiniMax-M3 |
|---|---|---|
| LMArena Expert | 1358 | 1461 |
| GPQA Diamond | — | 90.9% |
| Vectara Hallucination Rate | 12.3% | — |

## Multimodal

- Mercury 2: —
- MiniMax-M3: 40.2 (#51)

| Benchmark | Mercury 2 | MiniMax-M3 |
|---|---|---|
| LMArena Vision | — | 1253 |
| LMArena Document | — | 1435 |

## Multilingual

- Mercury 2: 46.6 (#157)
- MiniMax-M3: 53.0 (#75)

| Benchmark | Mercury 2 | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1331 | 1420 |
| LMArena Chinese | 1417 | 1463 |
| LMArena Russian | 1304 | 1428 |
| LMArena French | — | 1447 |
| LMArena German | — | 1426 |
| LMArena Japanese | — | 1381 |
| LMArena Korean | — | 1372 |
| LMArena Spanish | — | 1432 |

## Instruction Following

- Mercury 2: 70.2 (#165)
- MiniMax-M3: 75.5 (#62)

| Benchmark | Mercury 2 | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1329 | 1433 |

## Long Context

- Mercury 2: 40.5 (#154)
- MiniMax-M3: 44.2 (#72)

| Benchmark | Mercury 2 | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1330 | 1445 |

## Writing & Preference

- Mercury 2: 53.8 (#155)
- MiniMax-M3: 62.1 (#83)

| Benchmark | Mercury 2 | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1355 | 1433 |
| LMArena Creative Writing | 1289 | 1404 |
| LMArena Multi-Turn | 1358 | 1442 |
| EQ-Bench 4 | — | 1150 |

## FAQ

### Is Mercury 2 better than MiniMax-M3?

MiniMax-M3 is the stronger model overall, scoring 43.8 to 39.1 on the Noometry Index.

### Which is cheaper, Mercury 2 or MiniMax-M3?

Mercury 2 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.

### Is Mercury 2 or MiniMax-M3 better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 33.5 in the Noometry coding category.

### Which has the bigger context window?

MiniMax-M3 does, with 1M tokens against 128K.

### How many benchmarks do Mercury 2 and MiniMax-M3 share?

15 benchmarks have published results for both models. Mercury 2 has 17 scored results on Noometry and MiniMax-M3 has 41.