# Claude 3 Haiku vs GPT-4.1 mini

> GPT-4.1 mini is the stronger model overall, scoring 33.6 to 25.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-gpt-4-1-mini
- Last updated: 2026-10-12
- Shared benchmarks: 29

## Summary

- They share 29 benchmarks with published results for both. Claude 3 Haiku scores higher in 2 categories and GPT-4.1 mini in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4.1 mini leads 48.6 to 29.7.
- The biggest single-benchmark swing is MATH Level 5: 14.9% for Claude 3 Haiku and 87.3% for GPT-4.1 mini.

## Snapshot

| | Claude 3 Haiku | GPT-4.1 mini |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 25.9 | 33.6 |
| Rank | 340 | 240 |
| Context | — | 1.05M |
| Input $/M | — | $0.40 |
| Output $/M | — | $1.60 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3 Haiku: 26.4 (#325)
- GPT-4.1 mini: 30.6 (#293)

| Benchmark | Claude 3 Haiku | GPT-4.1 mini |
|---|---|---|
| WeirdML | 9.8% | 37.6% |
| BigCodeBench Instruct | 39.4% | 48.9% |
| LMArena Coding | 1199 | 1367 |
| CadEval | 12% | 16% |
| SWE-bench Verified (bash only) | — | 23.9% |
| Aider Polyglot | — | 32.4% |
| SciCode | — | 40.4% |
| BigCodeBench Complete | 50.1% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- GPT-4.1 mini: 33.3 (#55)

| Benchmark | Claude 3 Haiku | GPT-4.1 mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 50.5% |

## Reasoning

- Claude 3 Haiku: 16.3 (#307)
- GPT-4.1 mini: 10.8 (#340)

| Benchmark | Claude 3 Haiku | GPT-4.1 mini |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 48.6% |
| LMArena Hard Prompts | 1174 | 1349 |
| DTBench | 50.1% | 68.8% |
| LMCA | 8.8% | 21.1% |
| Epoch Capabilities Index | 118.35 | 135.01 |
| ARC-AGI-2 | — | 0% |
| ARC-AGI-1 | — | 3.5% |
| CritPt | — | 0% |
| Chess Puzzles | — | 7% |
| Mystery Game Puzzles | — | 7% |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.8 (#319)
- GPT-4.1 mini: 24.1 (#270)

| Benchmark | Claude 3 Haiku | GPT-4.1 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 44.7% |
| LMArena Math | 1188 | 1343 |
| MATH Level 5 | 14.9% | 87.3% |
| FrontierMath (Tiers 1-3) | — | 6.7% |
| Omni-MATH | — | 49.1% |
| FrontierMath (Feb 2025 set) | — | 4.5% |

## Knowledge

- Claude 3 Haiku: 17.3 (#285)
- GPT-4.1 mini: 34.7 (#194)

| Benchmark | Claude 3 Haiku | GPT-4.1 mini |
|---|---|---|
| GPQA Diamond | 36.3% | 65.8% |
| LMArena Expert | 1148 | 1338 |
| SimpleQA Verified | — | 12.7% |
| MMLU-Pro | — | 78.3% |
| Confabulations | 34.2% | — |
| GPQA (HELM) | — | 61.4% |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.6 (#128)
- GPT-4.1 mini: 35.8 (#82)

| Benchmark | Claude 3 Haiku | GPT-4.1 mini |
|---|---|---|
| LMArena Vision | 950 | 1181 |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.0 (#243)
- GPT-4.1 mini: 45.7 (#166)

| Benchmark | Claude 3 Haiku | GPT-4.1 mini |
|---|---|---|
| LMArena Non-English | 1178 | 1318 |
| LMArena Chinese | 1155 | 1329 |
| LMArena French | 1195 | 1358 |
| LMArena German | 1174 | 1351 |
| LMArena Japanese | 1102 | 1290 |
| LMArena Korean | 1109 | 1298 |
| LMArena Russian | 1204 | 1324 |
| LMArena Spanish | 1166 | 1319 |

## Instruction Following

- Claude 3 Haiku: 61.3 (#247)
- GPT-4.1 mini: 73.7 (#118)

| Benchmark | Claude 3 Haiku | GPT-4.1 mini |
|---|---|---|
| LMArena Instruction Following | 1173 | 1333 |
| IFEval | — | 90.4% |

## Long Context

- Claude 3 Haiku: 36.1 (#237)
- GPT-4.1 mini: 31.8 (#275)

| Benchmark | Claude 3 Haiku | GPT-4.1 mini |
|---|---|---|
| LMArena Longer Query | 1190 | 1344 |
| Fiction.LiveBench | — | 44.4% |

## Writing & Preference

- Claude 3 Haiku: 29.7 (#291)
- GPT-4.1 mini: 48.6 (#199)

| Benchmark | Claude 3 Haiku | GPT-4.1 mini |
|---|---|---|
| LMArena Text | 1195 | 1340 |
| LMArena Creative Writing | 1157 | 1300 |
| EQ-Bench Creative Writing | 717 | 1147 |
| LMArena Multi-Turn | 1190 | 1354 |
| WildBench | — | 83.8% |

## FAQ

### Is Claude 3 Haiku better than GPT-4.1 mini?

GPT-4.1 mini is the stronger model overall, scoring 33.6 to 25.9 on the Noometry Index.

### Is Claude 3 Haiku or GPT-4.1 mini better for coding?

GPT-4.1 mini scores higher on coding benchmarks: 30.6 versus 26.4 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and GPT-4.1 mini share?

29 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and GPT-4.1 mini has 47.