# Claude 3 Haiku vs MiniMax-M2.7

> MiniMax-M2.7 is the stronger model overall, scoring 37.7 to 25.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-minimax-m2-7
- Last updated: 2026-10-10
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and MiniMax-M2.7 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiniMax-M2.7 leads 58.9 to 29.7.
- The biggest single-benchmark swing is WeirdML: 9.8% for Claude 3 Haiku and 37% for MiniMax-M2.7.
- MiniMax-M2.7 has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Haiku | MiniMax-M2.7 |
|---|---|---|
| Provider | Anthropic | MiniMax |
| Noometry Index | 25.9 | 37.7 |
| Rank | 340 | 196 |
| Context | — | 205K |
| Input $/M | — | $0.30 |
| Output $/M | — | $1.20 |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Haiku: 26.4 (#325)
- MiniMax-M2.7: 41.8 (#120)

| Benchmark | Claude 3 Haiku | MiniMax-M2.7 |
|---|---|---|
| WeirdML | 9.8% | 37% |
| LMArena Coding | 1199 | 1454 |
| LMArena WebDev | — | 1398 |
| SciCode | — | 47% |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 599.25 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- MiniMax-M2.7: 25.1 (#111)

| Benchmark | Claude 3 Haiku | MiniMax-M2.7 |
|---|---|---|
| Terminal-Bench | — | 45.1% |
| ExploitBench | — | 13.3% |
| GBAEval | — | 0% |

## Reasoning

- Claude 3 Haiku: 16.3 (#307)
- MiniMax-M2.7: 19.7 (#253)

| Benchmark | Claude 3 Haiku | MiniMax-M2.7 |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1422 |
| Epoch Capabilities Index | 118.35 | 145.85 |
| Kagi LLM Benchmark | 34.2% | — |
| NYT Connections (extended) | — | 24.7% |
| CritPt | — | 0.6% |
| Thematic Generalization | — | 39.3% |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.8 (#319)
- MiniMax-M2.7: 25.9 (#263)

| Benchmark | Claude 3 Haiku | MiniMax-M2.7 |
|---|---|---|
| LMArena Math | 1188 | 1420 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| ProofBench | — | 3% |
| MATH Level 5 | 14.9% | — |

## Knowledge

- Claude 3 Haiku: 17.3 (#285)
- MiniMax-M2.7: 37.7 (#152)

| Benchmark | Claude 3 Haiku | MiniMax-M2.7 |
|---|---|---|
| LMArena Expert | 1148 | 1444 |
| GPQA Diamond | 36.3% | — |
| Confabulations | 34.2% | — |
| Vectara Hallucination Rate | — | 12.9% |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.6 (#128)
- MiniMax-M2.7: —

| Benchmark | Claude 3 Haiku | MiniMax-M2.7 |
|---|---|---|
| LMArena Vision | 950 | — |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.0 (#243)
- MiniMax-M2.7: 50.3 (#123)

| Benchmark | Claude 3 Haiku | MiniMax-M2.7 |
|---|---|---|
| LMArena Non-English | 1178 | 1382 |
| LMArena Chinese | 1155 | 1441 |
| LMArena French | 1195 | 1421 |
| LMArena German | 1174 | 1398 |
| LMArena Japanese | 1102 | 1262 |
| LMArena Korean | 1109 | 1313 |
| LMArena Russian | 1204 | 1383 |
| LMArena Spanish | 1166 | 1403 |

## Instruction Following

- Claude 3 Haiku: 61.3 (#247)
- MiniMax-M2.7: 74.1 (#103)

| Benchmark | Claude 3 Haiku | MiniMax-M2.7 |
|---|---|---|
| LMArena Instruction Following | 1173 | 1405 |

## Long Context

- Claude 3 Haiku: 36.1 (#237)
- MiniMax-M2.7: 43.3 (#99)

| Benchmark | Claude 3 Haiku | MiniMax-M2.7 |
|---|---|---|
| LMArena Longer Query | 1190 | 1419 |

## Writing & Preference

- Claude 3 Haiku: 29.7 (#291)
- MiniMax-M2.7: 58.9 (#112)

| Benchmark | Claude 3 Haiku | MiniMax-M2.7 |
|---|---|---|
| LMArena Text | 1195 | 1405 |
| LMArena Creative Writing | 1157 | 1354 |
| LMArena Multi-Turn | 1190 | 1412 |
| EQ-Bench Creative Writing | 717 | — |

## FAQ

### Is Claude 3 Haiku better than MiniMax-M2.7?

MiniMax-M2.7 is the stronger model overall, scoring 37.7 to 25.9 on the Noometry Index.

### Is Claude 3 Haiku or MiniMax-M2.7 better for coding?

MiniMax-M2.7 scores higher on coding benchmarks: 41.8 versus 26.4 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and MiniMax-M2.7 share?

19 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and MiniMax-M2.7 has 30.