Model comparison
Llama 2-70B vs Qwen2.5 72B Instruct
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 24.4 on the Noometry Index.
Last verified . 30 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 30 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and Qwen2.5 72B Instruct in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5 72B Instruct leads 27.0 to 7.4.
- The biggest single-benchmark swing is MATH Level 5: 3.3% for Llama 2-70B and 63.2% for Qwen2.5 72B Instruct.
Side by side
| Llama 2-70B | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 24.4 | 31.9 |
| 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 | 35 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 72B Instruct leads
Llama 2-70B: 31.4 (#286), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Llama 2-70B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Coding | 1079 | 1292 |
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Not comparable
Llama 2-70B: —, Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Llama 2-70B | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Qwen2.5 72B Instruct leads
Llama 2-70B: 14.4 (#325), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Llama 2-70B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1271 |
| DTBench | 41.6% | 62.9% |
| BIG-Bench Hard | 64.9% | 79.8% |
| Epoch Capabilities Index | 113.79 | 129 |
| ForecastBench | 51.4 | 57.5 |
| HellaSwag | 85.3% | 84.8% |
| PIQA | 82.8% | 82.6% |
| WinoGrande | 80.2% | 82.3% |
| LMCA | — | 13.4% |
| CommonsenseQA 2.0 | 50% | — |
| LAMBADA | 78.9% | — |
Math Qwen2.5 72B Instruct leads
Llama 2-70B: 8.1 (#326), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Llama 2-70B | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0% | 8.1% |
| LMArena Math | 1091 | 1283 |
| MATH Level 5 | 3.3% | 63.2% |
| Omni-MATH | — | 33% |
| GSM8K | 69.6% | — |
Knowledge Qwen2.5 72B Instruct leads
Llama 2-70B: 7.4 (#310), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Llama 2-70B | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 26.3% | 49.1% |
| LMArena Expert | 1039 | 1245 |
| ARC (AI2) Challenge | 78.3% | 94.5% |
| MMLU | 69.9% | 85.3% |
| TriviaQA | 87.6% | 71.9% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| BoolQ | 88.6% | — |
| OpenBookQA | 60.2% | — |
Multilingual Qwen2.5 72B Instruct leads
Llama 2-70B: 27.7 (#274), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Llama 2-70B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1045 | 1252 |
| LMArena Chinese | 995 | 1272 |
| LMArena French | 1090 | 1280 |
| LMArena German | 1041 | 1234 |
| LMArena Japanese | 927 | 1180 |
| LMArena Korean | 964 | 1188 |
| LMArena Russian | 1083 | 1264 |
| LMArena Spanish | 1143 | 1256 |
Instruction Following Qwen2.5 72B Instruct leads
Llama 2-70B: 54.9 (#278), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Llama 2-70B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1071 | 1254 |
| IFEval | — | 80.6% |
Long Context Qwen2.5 72B Instruct leads
Llama 2-70B: 32.3 (#270), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Llama 2-70B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1062 | 1282 |
Writing & Preference Qwen2.5 72B Instruct leads
Llama 2-70B: 32.3 (#279), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Llama 2-70B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1115 | 1269 |
| LMArena Creative Writing | 1075 | 1221 |
| LMArena Multi-Turn | 1088 | 1272 |
| WildBench | — | 80.2% |
Frequently asked questions
Is Llama 2-70B better than Qwen2.5 72B Instruct?
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 24.4 on the Noometry Index.
Is Llama 2-70B or Qwen2.5 72B Instruct better for coding?
Qwen2.5 72B Instruct scores higher on coding benchmarks: 33.2 versus 31.4 in the Noometry coding category.
How many benchmarks do Llama 2-70B and Qwen2.5 72B Instruct share?
30 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Qwen2.5 72B Instruct has 43.