Model comparison
Llama 13b vs Qwen2-72B
Qwen2-72B is the stronger model overall, scoring 30.0 to 24.4 on the Noometry Index.
Last verified . 10 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 10 benchmarks with published results for both. Llama 13b scores higher in 0 categories and Qwen2-72B in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2-72B leads 40.8 to 13.8.
Side by side
| Llama 13b | Qwen2-72B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 24.4 | 30.0 |
| Released | 2023-02-24 | 2024-06-07 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 21 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2-72B leads
Llama 13b: 21.4 (#337), Qwen2-72B: 29.1 (#310)
| Benchmark | Llama 13b | Qwen2-72B |
|---|---|---|
| LMArena Coding | 683 | 1196 |
| WeirdML | — | 11.3% |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
Agentic & Tool Use Not comparable
Llama 13b: —, Qwen2-72B: 17.0 (#146)
| Benchmark | Llama 13b | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Qwen2-72B leads
Llama 13b: 14.0 (#329), Qwen2-72B: 23.2 (#181)
| Benchmark | Llama 13b | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 728 | 1191 |
| Epoch Capabilities Index | 100.58 | 125.28 |
| BIG-Bench Hard | 37.9% | — |
| HellaSwag | 79.2% | — |
| LAMBADA | 75.2% | — |
| PIQA | 80.1% | — |
| WinoGrande | 73% | — |
Math Qwen2-72B leads
Llama 13b: 26.7 (#256), Qwen2-72B: 30.2 (#236)
| Benchmark | Llama 13b | Qwen2-72B |
|---|---|---|
| LMArena Math | 838 | 1235 |
| MATH Level 5 | — | 39.1% |
| GSM8K | 20.6% | — |
Knowledge Not comparable
Llama 13b: —, Qwen2-72B: 21.2 (#275)
| Benchmark | Llama 13b | Qwen2-72B |
|---|---|---|
| MMLU | 47.7% | 82.4% |
| GPQA Diamond | — | 40.8% |
| LMArena Expert | — | 1171 |
| ARC (AI2) Challenge | 52.7% | — |
| BoolQ | 78.7% | — |
| OpenBookQA | 56.4% | — |
| TriviaQA | 77.9% | — |
Multimodal Not comparable
Llama 13b: —, Qwen2-72B: —
| Benchmark | Llama 13b | Qwen2-72B |
|---|---|---|
| ScienceQA | 43.3% | — |
Multilingual Qwen2-72B leads
Llama 13b: 16.6 (#297), Qwen2-72B: 35.9 (#244)
| Benchmark | Llama 13b | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 819 | 1176 |
| LMArena Chinese | — | 1240 |
| LMArena French | — | 1170 |
| LMArena German | — | 1151 |
| LMArena Japanese | — | 1111 |
| LMArena Korean | — | 1083 |
| LMArena Russian | — | 1169 |
| LMArena Spanish | — | 1169 |
Instruction Following Qwen2-72B leads
Llama 13b: 36.7 (#305), Qwen2-72B: 61.7 (#241)
| Benchmark | Llama 13b | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 781 | 1181 |
Long Context Not comparable
Llama 13b: —, Qwen2-72B: 36.1 (#235)
| Benchmark | Llama 13b | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | — | 1192 |
Writing & Preference Qwen2-72B leads
Llama 13b: 13.8 (#312), Qwen2-72B: 40.8 (#241)
| Benchmark | Llama 13b | Qwen2-72B |
|---|---|---|
| LMArena Text | 834 | 1203 |
| LMArena Creative Writing | 794 | 1181 |
| LMArena Multi-Turn | 753 | 1196 |
Frequently asked questions
Is Llama 13b better than Qwen2-72B?
Qwen2-72B is the stronger model overall, scoring 30.0 to 24.4 on the Noometry Index.
Is Llama 13b or Qwen2-72B better for coding?
Qwen2-72B scores higher on coding benchmarks: 29.1 versus 21.4 in the Noometry coding category.
How many benchmarks do Llama 13b and Qwen2-72B share?
10 benchmarks have published results for both models. Llama 13b has 21 scored results on Noometry and Qwen2-72B has 26.