# Claude 3 Haiku vs MiniMax-M2.5

> MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 25.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-minimax-m2-5
- Last updated: 2026-10-10
- Shared benchmarks: 20

## Summary

- They share 20 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and MiniMax-M2.5 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiniMax-M2.5 leads 53.9 to 29.7.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 34.2% for Claude 3 Haiku and 55.2% for MiniMax-M2.5.
- MiniMax-M2.5 has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Haiku | MiniMax-M2.5 |
|---|---|---|
| Provider | Anthropic | MiniMax |
| Noometry Index | 25.9 | 38.3 |
| Rank | 340 | 188 |
| Context | — | 205K |
| Input $/M | — | $0.30 |
| Output $/M | — | $1.20 |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Haiku: 26.4 (#325)
- MiniMax-M2.5: 48.1 (#58)

| Benchmark | Claude 3 Haiku | MiniMax-M2.5 |
|---|---|---|
| LMArena Coding | 1199 | 1381 |
| SWE-bench Verified (bash only) | — | 75.8% |
| LMArena WebDev | — | 1387 |
| SWE-bench Multilingual | — | 68.3% |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 618.17 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- MiniMax-M2.5: 30.4 (#77)

| Benchmark | Claude 3 Haiku | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Vending-Bench 2 | — | -23.16 |

## Reasoning

- Claude 3 Haiku: 16.3 (#307)
- MiniMax-M2.5: 17.5 (#292)

| Benchmark | Claude 3 Haiku | MiniMax-M2.5 |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 55.2% |
| LMArena Hard Prompts | 1174 | 1372 |
| Epoch Capabilities Index | 118.35 | 146.68 |
| ARC-AGI-2 | — | 4.9% |
| NYT Connections (extended) | — | 16.8% |
| ARC-AGI-1 | — | 63.7% |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.8 (#319)
- MiniMax-M2.5: 26.9 (#253)

| Benchmark | Claude 3 Haiku | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1188 | 1378 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| ProofBench | — | 4% |
| MATH Level 5 | 14.9% | — |

## Knowledge

- Claude 3 Haiku: 17.3 (#285)
- MiniMax-M2.5: 39.2 (#135)

| Benchmark | Claude 3 Haiku | MiniMax-M2.5 |
|---|---|---|
| LMArena Expert | 1148 | 1379 |
| GPQA Diamond | 36.3% | — |
| Confabulations | 34.2% | — |
| Vectara Hallucination Rate | — | 9.1% |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.6 (#128)
- MiniMax-M2.5: —

| Benchmark | Claude 3 Haiku | MiniMax-M2.5 |
|---|---|---|
| LMArena Vision | 950 | — |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.0 (#243)
- MiniMax-M2.5: 47.1 (#152)

| Benchmark | Claude 3 Haiku | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1178 | 1338 |
| LMArena Chinese | 1155 | 1393 |
| LMArena French | 1195 | 1362 |
| LMArena German | 1174 | 1362 |
| LMArena Japanese | 1102 | 1171 |
| LMArena Korean | 1109 | 1232 |
| LMArena Russian | 1204 | 1358 |
| LMArena Spanish | 1166 | 1354 |

## Instruction Following

- Claude 3 Haiku: 61.3 (#247)
- MiniMax-M2.5: 71.5 (#148)

| Benchmark | Claude 3 Haiku | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1173 | 1353 |

## Long Context

- Claude 3 Haiku: 36.1 (#237)
- MiniMax-M2.5: 37.5 (#216)

| Benchmark | Claude 3 Haiku | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1190 | 1366 |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |

## Writing & Preference

- Claude 3 Haiku: 29.7 (#291)
- MiniMax-M2.5: 53.9 (#153)

| Benchmark | Claude 3 Haiku | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1195 | 1359 |
| LMArena Creative Writing | 1157 | 1331 |
| EQ-Bench Creative Writing | 717 | 1361 |
| LMArena Multi-Turn | 1190 | 1364 |

## FAQ

### Is Claude 3 Haiku better than MiniMax-M2.5?

MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 25.9 on the Noometry Index.

### Is Claude 3 Haiku or MiniMax-M2.5 better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 26.4 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and MiniMax-M2.5 share?

20 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and MiniMax-M2.5 has 33.