# Claude 3 Haiku vs Llama 3.2 1B

> Claude 3 Haiku is the stronger model overall, scoring 25.9 to 20.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-llama-3-2-1b
- Last updated: 2026-10-10
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. Claude 3 Haiku scores higher in 7 categories and Llama 3.2 1B in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Claude 3 Haiku leads 36.0 to 23.8.
- The biggest single-benchmark swing is BigCodeBench Complete: 50.1% for Claude 3 Haiku and 11.3% for Llama 3.2 1B.
- Llama 3.2 1B has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Haiku | Llama 3.2 1B |
|---|---|---|
| Provider | Anthropic | Meta |
| Noometry Index | 25.9 | 20.1 |
| Rank | 340 | 354 |
| Context | — | 60K |
| Input $/M | — | $0.027 |
| Output $/M | — | $0.20 |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Haiku: 26.4 (#325)
- Llama 3.2 1B: 21.1 (#338)

| Benchmark | Claude 3 Haiku | Llama 3.2 1B |
|---|---|---|
| BigCodeBench Instruct | 39.4% | 8.2% |
| LMArena Coding | 1199 | 1070 |
| BigCodeBench Complete | 50.1% | 11.3% |
| WeirdML | 9.8% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- Llama 3.2 1B: 14.6 (#150)

| Benchmark | Claude 3 Haiku | Llama 3.2 1B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 10.8% |
| BALROG | — | 6.6% |

## Reasoning

- Claude 3 Haiku: 16.3 (#307)
- Llama 3.2 1B: 16.2 (#308)

| Benchmark | Claude 3 Haiku | Llama 3.2 1B |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1044 |
| Epoch Capabilities Index | 118.35 | 101.99 |
| Kagi LLM Benchmark | 34.2% | — |
| Chess Puzzles | — | 0% |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.8 (#319)
- Llama 3.2 1B: 10.4 (#313)

| Benchmark | Claude 3 Haiku | Llama 3.2 1B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 0.6% |
| LMArena Math | 1188 | 1086 |
| MATH Level 5 | 14.9% | — |

## Knowledge

- Claude 3 Haiku: 17.3 (#285)
- Llama 3.2 1B: 7.2 (#312)

| Benchmark | Claude 3 Haiku | Llama 3.2 1B |
|---|---|---|
| GPQA Diamond | 36.3% | 23.9% |
| LMArena Expert | 1148 | 1007 |
| Confabulations | 34.2% | — |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.6 (#128)
- Llama 3.2 1B: —

| Benchmark | Claude 3 Haiku | Llama 3.2 1B |
|---|---|---|
| LMArena Vision | 950 | — |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.0 (#243)
- Llama 3.2 1B: 23.8 (#292)

| Benchmark | Claude 3 Haiku | Llama 3.2 1B |
|---|---|---|
| LMArena Non-English | 1178 | 973 |
| LMArena Chinese | 1155 | 959 |
| LMArena German | 1174 | 1014 |
| LMArena Russian | 1204 | 941 |
| LMArena French | 1195 | — |
| LMArena Japanese | 1102 | — |
| LMArena Korean | 1109 | — |
| LMArena Spanish | 1166 | — |

## Instruction Following

- Claude 3 Haiku: 61.3 (#247)
- Llama 3.2 1B: 52.4 (#290)

| Benchmark | Claude 3 Haiku | Llama 3.2 1B |
|---|---|---|
| LMArena Instruction Following | 1173 | 1031 |

## Long Context

- Claude 3 Haiku: 36.1 (#237)
- Llama 3.2 1B: 31.9 (#274)

| Benchmark | Claude 3 Haiku | Llama 3.2 1B |
|---|---|---|
| LMArena Longer Query | 1190 | 1050 |

## Writing & Preference

- Claude 3 Haiku: 29.7 (#291)
- Llama 3.2 1B: 21.3 (#310)

| Benchmark | Claude 3 Haiku | Llama 3.2 1B |
|---|---|---|
| LMArena Text | 1195 | 1055 |
| LMArena Creative Writing | 1157 | 1033 |
| EQ-Bench Creative Writing | 717 | 200 |
| LMArena Multi-Turn | 1190 | 1030 |

## FAQ

### Is Claude 3 Haiku better than Llama 3.2 1B?

Claude 3 Haiku is the stronger model overall, scoring 25.9 to 20.1 on the Noometry Index.

### Is Claude 3 Haiku or Llama 3.2 1B better for coding?

Claude 3 Haiku scores higher on coding benchmarks: 26.4 versus 21.1 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and Llama 3.2 1B share?

19 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Llama 3.2 1B has 22.