# Claude 2 vs Claude Haiku 4.5

> Claude Haiku 4.5 is the stronger model overall, scoring 39.3 to 24.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-2-vs-claude-haiku-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 5

## Summary

- They share 5 benchmarks with published results for both. Claude 2 scores higher in 1 category and Claude Haiku 4.5 in 2 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Haiku 4.5 leads 43.0 to 8.7.
- The biggest single-benchmark swing is MATH Level 5: 11.7% for Claude 2 and 96.4% for Claude Haiku 4.5.

## Snapshot

| | Claude 2 | Claude Haiku 4.5 |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 24.7 | 39.3 |
| Rank | 342 | 155 |
| Context | — | 200K |
| Input $/M | — | $1 |
| Output $/M | — | $5 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 2: —
- Claude Haiku 4.5: 43.9 (#71)

| Benchmark | Claude 2 | Claude Haiku 4.5 |
|---|---|---|
| SWE-bench Verified (bash only) | — | 66.6% |
| LMArena WebDev | — | 1330 |
| SWE-bench Multilingual | — | 64.7% |
| SciCode | — | 43.3% |
| WeirdML | — | 45.4% |
| LMArena Coding | — | 1454 |
| ALE-Bench | — | 653.48 |
| HumanEval+ | 61.6% | — |

## Agentic & Tool Use

- Claude 2: —
- Claude Haiku 4.5: 33.5 (#50)

| Benchmark | Claude 2 | Claude Haiku 4.5 |
|---|---|---|
| Terminal-Bench | — | 35.5% |
| Berkeley Function Calling Leaderboard | — | 68.7% |
| DeepResearch Bench | — | 45.5% |
| BALROG | — | 31.2% |
| ExploitBench | — | 13.7% |
| Vending-Bench 2 | — | 458.89 |

## Reasoning

- Claude 2: 20.8 (#213)
- Claude Haiku 4.5: 15.1 (#310)

| Benchmark | Claude 2 | Claude Haiku 4.5 |
|---|---|---|
| DTBench | 51.9% | 73.6% |
| Epoch Capabilities Index | 120.08 | 142.42 |
| ARC-AGI-2 | — | 4% |
| NYT Connections (extended) | — | 14.3% |
| ARC-AGI-1 | — | 47.7% |
| CritPt | — | 0% |
| Chess Puzzles | — | 8% |
| LMArena Hard Prompts | — | 1420 |
| LMCA | — | 30.9% |
| ForecastBench | — | 61.4 |

## Math

- Claude 2: 8.7 (#314)
- Claude Haiku 4.5: 43.0 (#69)

| Benchmark | Claude 2 | Claude Haiku 4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | 66.7% |
| MATH Level 5 | 11.7% | 96.4% |
| Omni-MATH | — | 56.1% |
| LMArena Math | — | 1395 |
| FrontierMath (Feb 2025 set) | — | 5.9% |
| FrontierMath Tier 4 (v1) | — | 2.1% |

## Knowledge

- Claude 2: 17.0 (#282)
- Claude Haiku 4.5: 37.6 (#146)

| Benchmark | Claude 2 | Claude Haiku 4.5 |
|---|---|---|
| GPQA Diamond | 34.7% | 71.2% |
| SimpleQA Verified | — | 13.2% |
| MMLU-Pro | — | 77.7% |
| Vectara Hallucination Rate | — | 9.8% |
| GPQA (HELM) | — | 60.5% |
| LMArena Expert | — | 1443 |
| MMLU | 78.5% | — |
| TriviaQA | 87.5% | — |

## Multimodal

- Claude 2: —
- Claude Haiku 4.5: 27.0 (#111)

| Benchmark | Claude 2 | Claude Haiku 4.5 |
|---|---|---|
| Blueprint-Bench 2 | — | 0% |
| LMArena Document | — | 1420 |

## Multilingual

- Claude 2: —
- Claude Haiku 4.5: 50.1 (#124)

| Benchmark | Claude 2 | Claude Haiku 4.5 |
|---|---|---|
| LMArena Non-English | — | 1376 |
| LMArena Chinese | — | 1415 |
| LMArena French | — | 1408 |
| LMArena German | — | 1377 |
| LMArena Japanese | — | 1339 |
| LMArena Korean | — | 1346 |
| LMArena Russian | — | 1380 |
| LMArena Spanish | — | 1420 |

## Instruction Following

- Claude 2: —
- Claude Haiku 4.5: 71.6 (#144)

| Benchmark | Claude 2 | Claude Haiku 4.5 |
|---|---|---|
| IFEval | — | 80.1% |
| LMArena Instruction Following | — | 1414 |

## Long Context

- Claude 2: —
- Claude Haiku 4.5: 43.6 (#86)

| Benchmark | Claude 2 | Claude Haiku 4.5 |
|---|---|---|
| LMArena Longer Query | — | 1427 |

## Writing & Preference

- Claude 2: —
- Claude Haiku 4.5: 58.1 (#119)

| Benchmark | Claude 2 | Claude Haiku 4.5 |
|---|---|---|
| LMArena Text | — | 1396 |
| LMArena Creative Writing | — | 1371 |
| WildBench | — | 83.9% |
| EQ-Bench 4 | — | 1064 |
| LMArena Multi-Turn | — | 1409 |

## FAQ

### Is Claude 2 better than Claude Haiku 4.5?

Claude Haiku 4.5 is the stronger model overall, scoring 39.3 to 24.7 on the Noometry Index.

### How many benchmarks do Claude 2 and Claude Haiku 4.5 share?

5 benchmarks have published results for both models. Claude 2 has 8 scored results on Noometry and Claude Haiku 4.5 has 53.