Model comparison
Llama 2-13B vs Qwen2.5 72B Instruct
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.5 to 29.1 on the Noometry Index.
Last verified . 26 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 26 benchmarks with published results for both. Llama 2-13B scores higher in 2 categories and Qwen2.5 72B Instruct in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2.5 72B Instruct leads 46.9 to 30.0.
- The biggest single-benchmark swing is DTBench: 42.2% for Llama 2-13B and 62.9% for Qwen2.5 72B Instruct.
Side by side
| Llama 2-13B | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 29.1 | 31.5 |
| Released | 2023-07-18 | 2024-09 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $5.60 |
| Results tracked | 32 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 72B Instruct leads
Llama 2-13B: 30.4 (#287), Qwen2.5 72B Instruct: 32.8 (#256)
| Benchmark | Llama 2-13B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Coding | 1062 | 1293 |
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Not comparable
Llama 2-13B: —, Qwen2.5 72B Instruct: 22.0 (#124)
| Benchmark | Llama 2-13B | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Qwen2.5 72B Instruct leads
Llama 2-13B: 12.3 (#333), Qwen2.5 72B Instruct: 21.4 (#197)
| Benchmark | Llama 2-13B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1051 | 1271 |
| DTBench | 42.2% | 62.9% |
| BIG-Bench Hard | 58.2% | 79.8% |
| Epoch Capabilities Index | 105.8 | 129.01 |
| HellaSwag | 80.7% | 84.8% |
| PIQA | 80.8% | 82.6% |
| WinoGrande | 72.8% | 82.3% |
| Chess Puzzles | 0% | — |
| LMCA | — | 13.4% |
| ForecastBench | — | 57.5 |
| LAMBADA | 76.5% | — |
Math Llama 2-13B leads
Llama 2-13B: 28.7 (#226), Qwen2.5 72B Instruct: 18.1 (#281)
| Benchmark | Llama 2-13B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Math | 1065 | 1283 |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| MATH Level 5 | — | 63.2% |
| GSM8K | 36.9% | — |
Knowledge Too close to call
Llama 2-13B: 27.5 (#244), Qwen2.5 72B Instruct: 26.9 (#247)
| Benchmark | Llama 2-13B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Expert | 1030 | 1245 |
| ARC (AI2) Challenge | 60.3% | 94.5% |
| MMLU | 55.6% | 85.3% |
| TriviaQA | 79.6% | 71.9% |
| GPQA Diamond | — | 49.1% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| BoolQ | 82.4% | — |
| OpenBookQA | 57% | — |
Multimodal Not comparable
Llama 2-13B: —, Qwen2.5 72B Instruct: —
| Benchmark | Llama 2-13B | Qwen2.5 72B Instruct |
|---|---|---|
| ScienceQA | 55.8% | — |
Multilingual Qwen2.5 72B Instruct leads
Llama 2-13B: 26.7 (#274), Qwen2.5 72B Instruct: 41.2 (#208)
| Benchmark | Llama 2-13B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1024 | 1252 |
| LMArena Chinese | 1001 | 1272 |
| LMArena French | 1043 | 1279 |
| LMArena German | 1009 | 1234 |
| LMArena Japanese | 894 | 1180 |
| LMArena Korean | 953 | 1188 |
| LMArena Russian | 1055 | 1264 |
| LMArena Spanish | 1086 | 1255 |
Instruction Following Qwen2.5 72B Instruct leads
Llama 2-13B: 53.5 (#282), Qwen2.5 72B Instruct: 65.7 (#216)
| Benchmark | Llama 2-13B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1046 | 1254 |
| IFEval | — | 80.6% |
Long Context Qwen2.5 72B Instruct leads
Llama 2-13B: 32.3 (#264), Qwen2.5 72B Instruct: 38.9 (#183)
| Benchmark | Llama 2-13B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1064 | 1282 |
Writing & Preference Qwen2.5 72B Instruct leads
Llama 2-13B: 30.0 (#284), Qwen2.5 72B Instruct: 46.9 (#210)
| Benchmark | Llama 2-13B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1085 | 1269 |
| LMArena Creative Writing | 1047 | 1221 |
| LMArena Multi-Turn | 1050 | 1272 |
| WildBench | — | 80.2% |
Frequently asked questions
Is Llama 2-13B better than Qwen2.5 72B Instruct?
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.5 to 29.1 on the Noometry Index.
Is Llama 2-13B or Qwen2.5 72B Instruct better for coding?
Qwen2.5 72B Instruct scores higher on coding benchmarks: 32.8 versus 30.4 in the Noometry coding category.
How many benchmarks do Llama 2-13B and Qwen2.5 72B Instruct share?
26 benchmarks have published results for both models. Llama 2-13B has 32 scored results on Noometry and Qwen2.5 72B Instruct has 43.