# Claude 3 Haiku vs Gemma 3 27B

> Gemma 3 27B is the stronger model overall, scoring 30.8 to 25.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-gemma-3-27b
- Last updated: 2026-10-12
- Shared benchmarks: 27

## Summary

- They share 27 benchmarks with published results for both. Claude 3 Haiku scores higher in 2 categories and Gemma 3 27B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 3 27B leads 52.5 to 29.7.
- The biggest single-benchmark swing is MATH Level 5: 14.9% for Claude 3 Haiku and 74% for Gemma 3 27B.
- Gemma 3 27B has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Haiku | Gemma 3 27B |
|---|---|---|
| Provider | Anthropic | Google |
| Noometry Index | 25.9 | 30.8 |
| Rank | 340 | 284 |
| Context | — | 131K |
| Input $/M | — | $0.08 |
| Output $/M | — | $0.16 |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Haiku: 26.4 (#325)
- Gemma 3 27B: 22.5 (#334)

| Benchmark | Claude 3 Haiku | Gemma 3 27B |
|---|---|---|
| LMArena Coding | 1199 | 1322 |
| Aider Polyglot | — | 4.9% |
| SciCode | — | 21.2% |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| LiveBench Coding | — | 39.9% |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- Gemma 3 27B: 25.1 (#110)

| Benchmark | Claude 3 Haiku | Gemma 3 27B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 29.5% |

## Reasoning

- Claude 3 Haiku: 16.3 (#307)
- Gemma 3 27B: 16.7 (#301)

| Benchmark | Claude 3 Haiku | Gemma 3 27B |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 40.4% |
| LMArena Hard Prompts | 1174 | 1340 |
| DTBench | 50.1% | 52.5% |
| LMCA | 8.8% | 12.3% |
| Epoch Capabilities Index | 118.35 | 130.04 |
| CritPt | — | 0% |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | — | 43.8% |
| LiveBench Data Analysis | — | 51.5% |
| ForecastBench | 53.2 | — |
| LiveBench | — | 50% |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.8 (#319)
- Gemma 3 27B: 25.9 (#265)

| Benchmark | Claude 3 Haiku | Gemma 3 27B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 22.5% |
| LMArena Math | 1188 | 1312 |
| MATH Level 5 | 14.9% | 74% |
| LiveBench Math | — | 55.4% |

## Knowledge

- Claude 3 Haiku: 17.3 (#285)
- Gemma 3 27B: 25.5 (#261)

| Benchmark | Claude 3 Haiku | Gemma 3 27B |
|---|---|---|
| GPQA Diamond | 36.3% | 47.7% |
| Confabulations | 34.2% | 40.3% |
| LMArena Expert | 1148 | 1304 |
| Vectara Hallucination Rate | — | 7.4% |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.6 (#128)
- Gemma 3 27B: 32.6 (#100)

| Benchmark | Claude 3 Haiku | Gemma 3 27B |
|---|---|---|
| LMArena Vision | 950 | 1164 |
| GeoBench | — | 52% |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.0 (#243)
- Gemma 3 27B: 46.9 (#155)

| Benchmark | Claude 3 Haiku | Gemma 3 27B |
|---|---|---|
| LMArena Non-English | 1178 | 1334 |
| LMArena Chinese | 1155 | 1346 |
| LMArena French | 1195 | 1368 |
| LMArena German | 1174 | 1362 |
| LMArena Japanese | 1102 | 1287 |
| LMArena Korean | 1109 | 1308 |
| LMArena Russian | 1204 | 1349 |
| LMArena Spanish | 1166 | 1349 |

## Instruction Following

- Claude 3 Haiku: 61.3 (#247)
- Gemma 3 27B: 70.6 (#160)

| Benchmark | Claude 3 Haiku | Gemma 3 27B |
|---|---|---|
| LMArena Instruction Following | 1173 | 1321 |
| LiveBench Instruction Following | — | 74.9% |

## Long Context

- Claude 3 Haiku: 36.1 (#237)
- Gemma 3 27B: 27.6 (#293)

| Benchmark | Claude 3 Haiku | Gemma 3 27B |
|---|---|---|
| LMArena Longer Query | 1190 | 1333 |
| Fiction.LiveBench | — | 33.3% |

## Writing & Preference

- Claude 3 Haiku: 29.7 (#291)
- Gemma 3 27B: 52.5 (#168)

| Benchmark | Claude 3 Haiku | Gemma 3 27B |
|---|---|---|
| LMArena Text | 1195 | 1358 |
| LMArena Creative Writing | 1157 | 1346 |
| EQ-Bench Creative Writing | 717 | 1266 |
| LMArena Multi-Turn | 1190 | 1345 |
| Short-Story Creative Writing | — | 79.9% |
| LiveBench Language | — | 34.6% |

## FAQ

### Is Claude 3 Haiku better than Gemma 3 27B?

Gemma 3 27B is the stronger model overall, scoring 30.8 to 25.9 on the Noometry Index.

### Is Claude 3 Haiku or Gemma 3 27B better for coding?

Claude 3 Haiku scores higher on coding benchmarks: 26.4 versus 22.5 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and Gemma 3 27B share?

27 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Gemma 3 27B has 43.