# Claude 3 Haiku vs Gemini 2.5 Pro

> Gemini 2.5 Pro is the stronger model overall, scoring 44.9 to 25.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-gemini-2-5-pro
- Last updated: 2026-10-10
- Shared benchmarks: 30

## Summary

- They share 30 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Gemini 2.5 Pro in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 2.5 Pro leads 55.8 to 17.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 84.7% for Gemini 2.5 Pro.

## Snapshot

| | Claude 3 Haiku | Gemini 2.5 Pro |
|---|---|---|
| Provider | Anthropic | Google |
| Noometry Index | 25.7 | 44.9 |
| Rank | 333 | 67 |
| Context | — | 1.05M |
| Input $/M | — | $1.25 |
| Output $/M | — | $10 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3 Haiku: 26.1 (#320)
- Gemini 2.5 Pro: 42.8 (#86)

| Benchmark | Claude 3 Haiku | Gemini 2.5 Pro |
|---|---|---|
| WeirdML | 9.8% | 54% |
| LMArena Coding | 1199 | 1452 |
| CadEval | 12% | 64% |
| SWE-bench Verified | — | 57.6% |
| SWE-bench Verified (bash only) | — | 53.6% |
| Aider Polyglot | — | 83.1% |
| LMArena WebDev | — | 1227 |
| SciCode | — | 42.8% |
| GSO | — | 3.9% |
| BigCodeBench Instruct | 39.4% | — |
| LiveBench Coding | — | 85.9% |
| BigCodeBench Complete | 50.1% | — |
| ALE-Bench | — | 785.52 |
| AlgoTune | — | 1.51 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- Gemini 2.5 Pro: 29.1 (#81)

| Benchmark | Claude 3 Haiku | Gemini 2.5 Pro |
|---|---|---|
| Terminal-Bench | — | 32.6% |
| GDPval | — | 23.3% |
| Remote Labor Index | — | 0.8% |
| TheAgentCompany | — | 30.3% |
| τ²-bench Banking | — | 13.7% |
| DeepResearch Bench | — | 42.8% |
| BALROG | — | 43.3% |
| LMArena Search | — | 1142 |
| METR Time Horizons | — | 55.4% |
| Vending-Bench 2 | — | 573.64 |

## Reasoning

- Claude 3 Haiku: 15.7 (#303)
- Gemini 2.5 Pro: 28.8 (#85)

| Benchmark | Claude 3 Haiku | Gemini 2.5 Pro |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 70.3% |
| LMArena Hard Prompts | 1174 | 1455 |
| DTBench | 50.1% | 82.4% |
| LMCA | 8.8% | 34.8% |
| Epoch Capabilities Index | 118.3 | 145.32 |
| ForecastBench | 53.2 | 61.3 |
| ARC-AGI-2 | — | 4.9% |
| SimpleBench | — | 62.4% |
| ARC-AGI-1 | — | 41% |
| CritPt | — | 2% |
| Chess Puzzles | — | 20% |
| EnigmaEval | — | 5.6% |
| LiveBench Reasoning | — | 89.8% |
| LiveBench Data Analysis | — | 79.9% |
| LiveBench | — | 82.3% |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.2 (#313)
- Gemini 2.5 Pro: 31.9 (#191)

| Benchmark | Claude 3 Haiku | Gemini 2.5 Pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 84.7% |
| LMArena Math | 1188 | 1450 |
| MATH Level 5 | 14.9% | 95.9% |
| FrontierMath (Tiers 1-3) | — | 24.6% |
| FrontierMath Tier 4 | — | 0% |
| Omni-MATH | — | 41.6% |
| LiveBench Math | — | 90.2% |
| FrontierMath (Feb 2025 set) | — | 14.1% |
| FrontierMath Tier 4 (v1) | — | 4.2% |

## Knowledge

- Claude 3 Haiku: 17.3 (#280)
- Gemini 2.5 Pro: 55.8 (#41)

| Benchmark | Claude 3 Haiku | Gemini 2.5 Pro |
|---|---|---|
| GPQA Diamond | 36.3% | 85.3% |
| Confabulations | 34.2% | 10.6% |
| LMArena Expert | 1148 | 1452 |
| Humanity's Last Exam | — | 21.6% |
| MMLU-Pro | — | 86.3% |
| Vectara Hallucination Rate | — | 7% |
| GPQA (HELM) | — | 74.9% |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.1 (#123)
- Gemini 2.5 Pro: 44.5 (#16)

| Benchmark | Claude 3 Haiku | Gemini 2.5 Pro |
|---|---|---|
| LMArena Vision | 950 | 1263 |
| GeoBench | — | 86% |
| VPCT | — | 48% |
| LMArena Document | — | 1421 |
| ScienceQA | 72% | — |
| SpatialViz-Bench | — | 44.7% |

## Multilingual

- Claude 3 Haiku: 36.1 (#238)
- Gemini 2.5 Pro: 55.5 (#30)

| Benchmark | Claude 3 Haiku | Gemini 2.5 Pro |
|---|---|---|
| LMArena Non-English | 1178 | 1452 |
| LMArena Chinese | 1155 | 1507 |
| LMArena French | 1194 | 1473 |
| LMArena German | 1174 | 1487 |
| LMArena Japanese | 1102 | 1462 |
| LMArena Korean | 1108 | 1435 |
| LMArena Russian | 1204 | 1460 |
| LMArena Spanish | 1165 | 1473 |

## Instruction Following

- Claude 3 Haiku: 61.5 (#242)
- Gemini 2.5 Pro: 75.2 (#71)

| Benchmark | Claude 3 Haiku | Gemini 2.5 Pro |
|---|---|---|
| LMArena Instruction Following | 1174 | 1436 |
| LiveBench Instruction Following | — | 80.6% |
| IFEval | — | 84% |

## Long Context

- Claude 3 Haiku: 36.1 (#232)
- Gemini 2.5 Pro: 59.7 (#5)

| Benchmark | Claude 3 Haiku | Gemini 2.5 Pro |
|---|---|---|
| LMArena Longer Query | 1191 | 1449 |
| Fiction.LiveBench | — | 91.7% |

## Writing & Preference

- Claude 3 Haiku: 29.8 (#286)
- Gemini 2.5 Pro: 63.9 (#60)

| Benchmark | Claude 3 Haiku | Gemini 2.5 Pro |
|---|---|---|
| LMArena Text | 1195 | 1458 |
| LMArena Creative Writing | 1157 | 1453 |
| EQ-Bench Creative Writing | 717 | 1421 |
| LMArena Multi-Turn | 1190 | 1453 |
| Short-Story Creative Writing | — | 83.8% |
| WildBench | — | 85.7% |
| LiveBench Language | — | 67.8% |

## FAQ

### Is Claude 3 Haiku better than Gemini 2.5 Pro?

Gemini 2.5 Pro is the stronger model overall, scoring 44.9 to 25.7 on the Noometry Index.

### Is Claude 3 Haiku or Gemini 2.5 Pro better for coding?

Gemini 2.5 Pro scores higher on coding benchmarks: 42.8 versus 26.1 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and Gemini 2.5 Pro share?

30 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Gemini 2.5 Pro has 78.