# Claude 3 Haiku vs MiniMax-M3

> MiniMax-M3 is the stronger model overall, scoring 43.5 to 25.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-minimax-m3
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and MiniMax-M3 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 17.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 71.1% for MiniMax-M3.
- MiniMax-M3 has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Haiku | MiniMax-M3 |
|---|---|---|
| Provider | Anthropic | MiniMax |
| Noometry Index | 25.7 | 43.5 |
| Rank | 333 | 81 |
| Context | — | 1M |
| Input $/M | — | $0.30 |
| Output $/M | — | $1.20 |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Haiku: 26.1 (#320)
- MiniMax-M3: 41.5 (#111)

| Benchmark | Claude 3 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Coding | 1199 | 1470 |
| FrontierCode | — | 14.7% |
| LMArena WebDev | — | 1482 |
| SciCode | — | 47.1% |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 640.02 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- MiniMax-M3: 22.5 (#121)

| Benchmark | Claude 3 Haiku | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |

## Reasoning

- Claude 3 Haiku: 15.7 (#303)
- MiniMax-M3: 29.3 (#83)

| Benchmark | Claude 3 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1447 |
| DTBench | 50.1% | 78.9% |
| LMCA | 8.8% | 33.7% |
| Epoch Capabilities Index | 118.3 | 147 |
| ForecastBench | 53.2 | 61.4 |
| SimpleBench | — | 45.8% |
| Kagi LLM Benchmark | 34.2% | — |
| NYT Connections (extended) | — | 65.1% |
| CritPt | — | 3.7% |
| Chess Puzzles | — | 14% |
| Mystery Game Puzzles | — | 8% |
| Surface Evolver Bench | — | 55% |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.2 (#313)
- MiniMax-M3: 39.1 (#83)

| Benchmark | Claude 3 Haiku | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 71.1% |
| LMArena Math | 1188 | 1432 |
| ProofBench | — | 18% |
| MATH Level 5 | 14.9% | — |

## Knowledge

- Claude 3 Haiku: 17.3 (#280)
- MiniMax-M3: 58.4 (#32)

| Benchmark | Claude 3 Haiku | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 36.3% | 90.9% |
| LMArena Expert | 1148 | 1462 |
| Confabulations | 34.2% | — |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.1 (#123)
- MiniMax-M3: 39.5 (#45)

| Benchmark | Claude 3 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 950 | 1254 |
| LMArena Document | — | 1435 |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.1 (#238)
- MiniMax-M3: 53.2 (#70)

| Benchmark | Claude 3 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1178 | 1420 |
| LMArena Chinese | 1155 | 1465 |
| LMArena French | 1194 | 1445 |
| LMArena German | 1174 | 1424 |
| LMArena Japanese | 1102 | 1381 |
| LMArena Korean | 1108 | 1372 |
| LMArena Russian | 1204 | 1430 |
| LMArena Spanish | 1165 | 1431 |

## Instruction Following

- Claude 3 Haiku: 61.5 (#242)
- MiniMax-M3: 75.7 (#56)

| Benchmark | Claude 3 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1174 | 1433 |

## Long Context

- Claude 3 Haiku: 36.1 (#232)
- MiniMax-M3: 44.2 (#68)

| Benchmark | Claude 3 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1191 | 1445 |

## Writing & Preference

- Claude 3 Haiku: 29.8 (#286)
- MiniMax-M3: 62.3 (#79)

| Benchmark | Claude 3 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1195 | 1433 |
| LMArena Creative Writing | 1157 | 1405 |
| LMArena Multi-Turn | 1190 | 1441 |
| EQ-Bench Creative Writing | 717 | — |
| EQ-Bench 4 | — | 1150 |

## FAQ

### Is Claude 3 Haiku better than MiniMax-M3?

MiniMax-M3 is the stronger model overall, scoring 43.5 to 25.7 on the Noometry Index.

### Is Claude 3 Haiku or MiniMax-M3 better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 26.1 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and MiniMax-M3 share?

24 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and MiniMax-M3 has 41.