# Claude 3.5 Haiku vs GPT-4.1

> GPT-4.1 is the stronger model overall, scoring 35.9 to 29.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-5-haiku-vs-gpt-4-1
- Last updated: 2026-10-10
- Shared benchmarks: 34

## Summary

- They share 34 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 1 category and GPT-4.1 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-4.1 leads 37.1 to 18.7.
- The biggest single-benchmark swing is GeoBench: 34% for Claude 3.5 Haiku and 72% for GPT-4.1.

## Snapshot

| | Claude 3.5 Haiku | GPT-4.1 |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 29.2 | 35.9 |
| Rank | 315 | 219 |
| Context | — | 1.05M |
| Input $/M | — | $2 |
| Output $/M | — | $8 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3.5 Haiku: 32.9 (#265)
- GPT-4.1: 34.4 (#238)

| Benchmark | Claude 3.5 Haiku | GPT-4.1 |
|---|---|---|
| Aider Polyglot | 28% | 52.4% |
| WeirdML | 30.7% | 39% |
| LMArena Coding | 1286 | 1391 |
| CadEval | 32% | 42% |
| SWE-bench Verified | — | 48.5% |
| SWE-bench Verified (bash only) | — | 39.6% |
| SciCode | 27.4% | — |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| ALE-Bench | — | 558.1 |

## Agentic & Tool Use

- Claude 3.5 Haiku: 28.0 (#95)
- GPT-4.1: 34.7 (#43)

| Benchmark | Claude 3.5 Haiku | GPT-4.1 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 54% |
| BALROG | 19.3% | — |

## Reasoning

- Claude 3.5 Haiku: 17.7 (#290)
- GPT-4.1: 11.7 (#339)

| Benchmark | Claude 3.5 Haiku | GPT-4.1 |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1384 |
| DTBench | 56.7% | 68.3% |
| Epoch Capabilities Index | 127.15 | 136.78 |
| ARC-AGI-2 | — | 0.4% |
| SimpleBench | — | 27% |
| Kagi LLM Benchmark | — | 52.3% |
| ARC-AGI-1 | — | 5.5% |
| CritPt | 0% | — |
| Chess Puzzles | — | 6% |
| EnigmaEval | — | 2.2% |
| LiveBench Reasoning | 28.1% | — |
| LiveBench Data Analysis | 48.5% | — |
| LMCA | — | 25.6% |
| ForecastBench | — | 61.5 |
| LiveBench | 43.5% | — |

## Math

- Claude 3.5 Haiku: 14.7 (#300)
- GPT-4.1: 22.3 (#280)

| Benchmark | Claude 3.5 Haiku | GPT-4.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 38.3% |
| Omni-MATH | 22.4% | 47.1% |
| LMArena Math | 1244 | 1370 |
| MATH Level 5 | 46.4% | 83% |
| FrontierMath (Feb 2025 set) | 0.3% | 5.5% |
| FrontierMath (Tiers 1-3) | — | 6% |
| LiveBench Math | 35.5% | — |
| FrontierMath Tier 4 (v1) | — | 0% |

## Knowledge

- Claude 3.5 Haiku: 18.7 (#281)
- GPT-4.1: 37.1 (#160)

| Benchmark | Claude 3.5 Haiku | GPT-4.1 |
|---|---|---|
| GPQA Diamond | 38.1% | 66.9% |
| MMLU-Pro | 60.5% | 81.1% |
| GPQA (HELM) | 36.3% | 65.9% |
| LMArena Expert | 1208 | 1364 |
| Humanity's Last Exam | — | 5.4% |
| SimpleQA Verified | — | 31.1% |
| Confabulations | 36.7% | — |
| Vectara Hallucination Rate | — | 5.6% |
| MMLU | 74.3% | — |

## Multimodal

- Claude 3.5 Haiku: 26.8 (#117)
- GPT-4.1: 38.2 (#67)

| Benchmark | Claude 3.5 Haiku | GPT-4.1 |
|---|---|---|
| LMArena Vision | 1092 | 1211 |
| GeoBench | 34% | 72% |

## Multilingual

- Claude 3.5 Haiku: 40.0 (#218)
- GPT-4.1: 49.4 (#133)

| Benchmark | Claude 3.5 Haiku | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1238 | 1370 |
| LMArena Chinese | 1229 | 1382 |
| LMArena French | 1264 | 1382 |
| LMArena German | 1237 | 1381 |
| LMArena Japanese | 1175 | 1319 |
| LMArena Korean | 1173 | 1339 |
| LMArena Russian | 1253 | 1377 |
| LMArena Spanish | 1261 | 1376 |

## Instruction Following

- Claude 3.5 Haiku: 62.9 (#234)
- GPT-4.1: 71.3 (#153)

| Benchmark | Claude 3.5 Haiku | GPT-4.1 |
|---|---|---|
| IFEval | 79.2% | 83.8% |
| LMArena Instruction Following | 1241 | 1367 |
| LiveBench Instruction Following | 61.9% | — |

## Long Context

- Claude 3.5 Haiku: 38.3 (#200)
- GPT-4.1: 40.0 (#163)

| Benchmark | Claude 3.5 Haiku | GPT-4.1 |
|---|---|---|
| LMArena Longer Query | 1261 | 1385 |
| Fiction.LiveBench | — | 63.9% |

## Writing & Preference

- Claude 3.5 Haiku: 42.7 (#234)
- GPT-4.1: 57.6 (#125)

| Benchmark | Claude 3.5 Haiku | GPT-4.1 |
|---|---|---|
| LMArena Text | 1255 | 1383 |
| LMArena Creative Writing | 1233 | 1363 |
| EQ-Bench Creative Writing | 1146 | 1420 |
| WildBench | 76% | 85.4% |
| LMArena Multi-Turn | 1265 | 1398 |
| Short-Story Creative Writing | 73.5% | — |
| LiveBench Language | 35.4% | — |

## FAQ

### Is Claude 3.5 Haiku better than GPT-4.1?

GPT-4.1 is the stronger model overall, scoring 35.9 to 29.2 on the Noometry Index.

### Is Claude 3.5 Haiku or GPT-4.1 better for coding?

GPT-4.1 scores higher on coding benchmarks: 34.4 versus 32.9 in the Noometry coding category.

### How many benchmarks do Claude 3.5 Haiku and GPT-4.1 share?

34 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and GPT-4.1 has 52.