Model comparison
Llama 3.2 3B vs Qwen2-72B
Qwen2-72B is the stronger model overall, scoring 29.5 to 28.4 on the Noometry Index.
Last verified . 15 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 15 benchmarks with published results for both. Llama 3.2 3B scores higher in 3 categories and Qwen2-72B in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2-72B leads 41.0 to 24.9.
- The biggest single-benchmark swing is BigCodeBench Complete: 28.3% for Llama 3.2 3B and 54% for Qwen2-72B.
Side by side
| Llama 3.2 3B | Qwen2-72B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 28.4 | 29.5 |
| Released | 2024-09-24 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 118K | — |
| Input $ / M tokens | $0.05 | — |
| Output $ / M tokens | $0.33 | — |
| Results tracked | 18 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2-72B leads
Llama 3.2 3B: 27.1 (#314), Qwen2-72B: 28.7 (#303)
| Benchmark | Llama 3.2 3B | Qwen2-72B |
|---|---|---|
| BigCodeBench Instruct | 23.4% | 38.5% |
| LMArena Coding | 1098 | 1196 |
| BigCodeBench Complete | 28.3% | 54% |
| WeirdML | — | 11.3% |
Agentic & Tool Use Llama 3.2 3B leads
Llama 3.2 3B: 19.9 (#135), Qwen2-72B: 16.9 (#138)
| Benchmark | Llama 3.2 3B | Qwen2-72B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 21.9% | — |
| TheAgentCompany | — | 1.1% |
| BALROG | 10.1% | — |
| METR Time Horizons | — | 29.9% |
Reasoning Qwen2-72B leads
Llama 3.2 3B: 20.1 (#225), Qwen2-72B: 22.2 (#184)
| Benchmark | Llama 3.2 3B | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1095 | 1191 |
| Epoch Capabilities Index | — | 125.27 |
Math Llama 3.2 3B leads
Llama 3.2 3B: 29.9 (#213), Qwen2-72B: 27.9 (#232)
| Benchmark | Llama 3.2 3B | Qwen2-72B |
|---|---|---|
| LMArena Math | 1126 | 1235 |
| MATH Level 5 | — | 39.1% |
Knowledge Llama 3.2 3B leads
Llama 3.2 3B: 29.1 (#232), Qwen2-72B: 21.2 (#270)
| Benchmark | Llama 3.2 3B | Qwen2-72B |
|---|---|---|
| LMArena Expert | 1090 | 1171 |
| GPQA Diamond | — | 40.8% |
| MMLU | — | 82.4% |
Multilingual Qwen2-72B leads
Llama 3.2 3B: 26.4 (#276), Qwen2-72B: 36.0 (#239)
| Benchmark | Llama 3.2 3B | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1019 | 1176 |
| LMArena Chinese | 1017 | 1240 |
| LMArena German | 1056 | 1151 |
| LMArena Russian | 949 | 1169 |
| LMArena French | — | 1169 |
| LMArena Japanese | — | 1111 |
| LMArena Korean | — | 1083 |
| LMArena Spanish | — | 1168 |
Instruction Following Qwen2-72B leads
Llama 3.2 3B: 56.3 (#270), Qwen2-72B: 62.0 (#236)
| Benchmark | Llama 3.2 3B | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1089 | 1181 |
Long Context Qwen2-72B leads
Llama 3.2 3B: 33.4 (#256), Qwen2-72B: 36.1 (#230)
| Benchmark | Llama 3.2 3B | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1101 | 1193 |
Writing & Preference Qwen2-72B leads
Llama 3.2 3B: 24.9 (#302), Qwen2-72B: 41.0 (#236)
| Benchmark | Llama 3.2 3B | Qwen2-72B |
|---|---|---|
| LMArena Text | 1110 | 1203 |
| LMArena Creative Writing | 1094 | 1181 |
| LMArena Multi-Turn | 1105 | 1195 |
| EQ-Bench Creative Writing | 595 | — |
Frequently asked questions
Is Llama 3.2 3B better than Qwen2-72B?
Qwen2-72B is the stronger model overall, scoring 29.5 to 28.4 on the Noometry Index.
Is Llama 3.2 3B or Qwen2-72B better for coding?
Qwen2-72B scores higher on coding benchmarks: 28.7 versus 27.1 in the Noometry coding category.
How many benchmarks do Llama 3.2 3B and Qwen2-72B share?
15 benchmarks have published results for both models. Llama 3.2 3B has 18 scored results on Noometry and Qwen2-72B has 26.