# Llama 3.2 1B vs Qwen2-72B

> Qwen2-72B is the stronger model overall, scoring 30.0 to 20.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/llama-3-2-1b-vs-qwen2-72b
- Last updated: 2026-10-12
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Llama 3.2 1B scores higher in 0 categories and Qwen2-72B in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2-72B leads 30.2 to 10.4.
- The biggest single-benchmark swing is BigCodeBench Complete: 11.3% for Llama 3.2 1B and 54% for Qwen2-72B.

## Snapshot

| | Llama 3.2 1B | Qwen2-72B |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 20.1 | 30.0 |
| Rank | 354 | 300 |
| Context | 60K | — |
| Input $/M | $0.027 | — |
| Output $/M | $0.20 | — |
| Weights | Open | Open |

## Coding

- Llama 3.2 1B: 21.1 (#338)
- Qwen2-72B: 29.1 (#310)

| Benchmark | Llama 3.2 1B | Qwen2-72B |
|---|---|---|
| BigCodeBench Instruct | 8.2% | 38.5% |
| LMArena Coding | 1070 | 1196 |
| BigCodeBench Complete | 11.3% | 54% |
| WeirdML | — | 11.3% |

## Agentic & Tool Use

- Llama 3.2 1B: 14.6 (#150)
- Qwen2-72B: 17.0 (#146)

| Benchmark | Llama 3.2 1B | Qwen2-72B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 10.8% | — |
| TheAgentCompany | — | 1.1% |
| BALROG | 6.6% | — |
| METR Time Horizons | — | 29.9% |

## Reasoning

- Llama 3.2 1B: 16.2 (#308)
- Qwen2-72B: 23.2 (#181)

| Benchmark | Llama 3.2 1B | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1044 | 1191 |
| Epoch Capabilities Index | 101.99 | 125.28 |
| Chess Puzzles | 0% | — |

## Math

- Llama 3.2 1B: 10.4 (#313)
- Qwen2-72B: 30.2 (#236)

| Benchmark | Llama 3.2 1B | Qwen2-72B |
|---|---|---|
| LMArena Math | 1086 | 1235 |
| OTIS Mock AIME 2024-2025 | 0.6% | — |
| MATH Level 5 | — | 39.1% |

## Knowledge

- Llama 3.2 1B: 7.2 (#312)
- Qwen2-72B: 21.2 (#275)

| Benchmark | Llama 3.2 1B | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 23.9% | 40.8% |
| LMArena Expert | 1007 | 1171 |
| MMLU | — | 82.4% |

## Multilingual

- Llama 3.2 1B: 23.8 (#292)
- Qwen2-72B: 35.9 (#244)

| Benchmark | Llama 3.2 1B | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 973 | 1176 |
| LMArena Chinese | 959 | 1240 |
| LMArena German | 1014 | 1151 |
| LMArena Russian | 941 | 1169 |
| LMArena French | — | 1170 |
| LMArena Japanese | — | 1111 |
| LMArena Korean | — | 1083 |
| LMArena Spanish | — | 1169 |

## Instruction Following

- Llama 3.2 1B: 52.4 (#290)
- Qwen2-72B: 61.7 (#241)

| Benchmark | Llama 3.2 1B | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1031 | 1181 |

## Long Context

- Llama 3.2 1B: 31.9 (#274)
- Qwen2-72B: 36.1 (#235)

| Benchmark | Llama 3.2 1B | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1050 | 1192 |

## Writing & Preference

- Llama 3.2 1B: 21.3 (#310)
- Qwen2-72B: 40.8 (#241)

| Benchmark | Llama 3.2 1B | Qwen2-72B |
|---|---|---|
| LMArena Text | 1055 | 1203 |
| LMArena Creative Writing | 1033 | 1181 |
| LMArena Multi-Turn | 1030 | 1196 |
| EQ-Bench Creative Writing | 200 | — |

## FAQ

### Is Llama 3.2 1B better than Qwen2-72B?

Qwen2-72B is the stronger model overall, scoring 30.0 to 20.1 on the Noometry Index.

### Is Llama 3.2 1B or Qwen2-72B better for coding?

Qwen2-72B scores higher on coding benchmarks: 29.1 versus 21.1 in the Noometry coding category.

### How many benchmarks do Llama 3.2 1B and Qwen2-72B share?

17 benchmarks have published results for both models. Llama 3.2 1B has 22 scored results on Noometry and Qwen2-72B has 26.