# Claude 3 Haiku vs MiniMax-M2

> MiniMax-M2 is the stronger model overall, scoring 36.9 to 25.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-minimax-m2
- Last updated: 2026-10-10
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and MiniMax-M2 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where MiniMax-M2 leads 34.6 to 9.2.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 34.2% for Claude 3 Haiku and 57.8% for MiniMax-M2.
- MiniMax-M2 has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Haiku | MiniMax-M2 |
|---|---|---|
| Provider | Anthropic | MiniMax |
| Noometry Index | 25.7 | 36.9 |
| Rank | 333 | 201 |
| Context | — | 205K |
| Input $/M | — | $0.30 |
| Output $/M | — | $1.20 |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Haiku: 26.1 (#320)
- MiniMax-M2: 39.1 (#151)

| Benchmark | Claude 3 Haiku | MiniMax-M2 |
|---|---|---|
| LMArena Coding | 1199 | 1370 |
| SWE-bench Verified (bash only) | — | 61% |
| LMArena WebDev | — | 1297 |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- MiniMax-M2: 24.9 (#102)

| Benchmark | Claude 3 Haiku | MiniMax-M2 |
|---|---|---|
| Terminal-Bench | — | 30% |
| Vending-Bench 2 | — | 160.6 |

## Reasoning

- Claude 3 Haiku: 15.7 (#303)
- MiniMax-M2: 18.6 (#256)

| Benchmark | Claude 3 Haiku | MiniMax-M2 |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 57.8% |
| LMArena Hard Prompts | 1174 | 1357 |
| NYT Connections (extended) | — | 14.8% |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| Epoch Capabilities Index | 118.3 | — |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.2 (#313)
- MiniMax-M2: 34.6 (#158)

| Benchmark | Claude 3 Haiku | MiniMax-M2 |
|---|---|---|
| LMArena Math | 1188 | 1350 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| MATH Level 5 | 14.9% | — |

## Knowledge

- Claude 3 Haiku: 17.3 (#280)
- MiniMax-M2: 36.3 (#160)

| Benchmark | Claude 3 Haiku | MiniMax-M2 |
|---|---|---|
| LMArena Expert | 1148 | 1336 |
| GPQA Diamond | 36.3% | — |
| Confabulations | 34.2% | — |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.1 (#123)
- MiniMax-M2: —

| Benchmark | Claude 3 Haiku | MiniMax-M2 |
|---|---|---|
| LMArena Vision | 950 | — |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.1 (#238)
- MiniMax-M2: 45.6 (#165)

| Benchmark | Claude 3 Haiku | MiniMax-M2 |
|---|---|---|
| LMArena Non-English | 1178 | 1314 |
| LMArena Chinese | 1155 | 1367 |
| LMArena French | 1194 | 1336 |
| LMArena German | 1174 | 1366 |
| LMArena Russian | 1204 | 1332 |
| LMArena Spanish | 1165 | 1326 |
| LMArena Japanese | 1102 | — |
| LMArena Korean | 1108 | — |

## Instruction Following

- Claude 3 Haiku: 61.5 (#242)
- MiniMax-M2: 70.4 (#160)

| Benchmark | Claude 3 Haiku | MiniMax-M2 |
|---|---|---|
| LMArena Instruction Following | 1174 | 1329 |

## Long Context

- Claude 3 Haiku: 36.1 (#232)
- MiniMax-M2: 40.5 (#148)

| Benchmark | Claude 3 Haiku | MiniMax-M2 |
|---|---|---|
| LMArena Longer Query | 1191 | 1332 |

## Writing & Preference

- Claude 3 Haiku: 29.8 (#286)
- MiniMax-M2: 53.3 (#157)

| Benchmark | Claude 3 Haiku | MiniMax-M2 |
|---|---|---|
| LMArena Text | 1195 | 1340 |
| LMArena Creative Writing | 1157 | 1288 |
| LMArena Multi-Turn | 1190 | 1362 |
| EQ-Bench Creative Writing | 717 | — |

## FAQ

### Is Claude 3 Haiku better than MiniMax-M2?

MiniMax-M2 is the stronger model overall, scoring 36.9 to 25.7 on the Noometry Index.

### Is Claude 3 Haiku or MiniMax-M2 better for coding?

MiniMax-M2 scores higher on coding benchmarks: 39.1 versus 26.1 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and MiniMax-M2 share?

16 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and MiniMax-M2 has 21.