Model comparison
Llama 3-70B vs Qwen1.5-72B
Qwen1.5-72B is the stronger model overall, scoring 30.3 to 28.5 on the Noometry Index.
Last verified . 22 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. Llama 3-70B scores higher in 6 categories and Qwen1.5-72B in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen1.5-72B leads 30.7 to 12.0.
- The biggest single-benchmark swing is BigCodeBench Complete: 54.5% for Llama 3-70B and 40.3% for Qwen1.5-72B.
Side by side
| Llama 3-70B | Qwen1.5-72B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 28.5 | 30.3 |
| Released | 2024-04-18 | 2024-02-04 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 31 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3-70B leads
Llama 3-70B: 35.3 (#215), Qwen1.5-72B: 31.4 (#272)
| Benchmark | Llama 3-70B | Qwen1.5-72B |
|---|---|---|
| BigCodeBench Instruct | 43.6% | 33.2% |
| LMArena Coding | 1206 | 1166 |
| BigCodeBench Complete | 54.5% | 40.3% |
| HumanEval+ | 72% | 59.1% |
| MBPP+ | 69% | 61.6% |
Agentic & Tool Use Not comparable
Llama 3-70B: 20.9 (#132), Qwen1.5-72B: —
| Benchmark | Llama 3-70B | Qwen1.5-72B |
|---|---|---|
| Cybench | 5% | — |
Reasoning Qwen1.5-72B leads
Llama 3-70B: 17.4 (#284), Qwen1.5-72B: 21.2 (#201)
| Benchmark | Llama 3-70B | Qwen1.5-72B |
|---|---|---|
| LMArena Hard Prompts | 1195 | 1148 |
| Kagi LLM Benchmark | 35.1% | — |
| DTBench | 54.2% | — |
| Epoch Capabilities Index | 122.91 | — |
| ForecastBench | 57.1 | — |
| WinoGrande | 83.5% | — |
Math Qwen1.5-72B leads
Llama 3-70B: 12.0 (#299), Qwen1.5-72B: 30.7 (#205)
| Benchmark | Llama 3-70B | Qwen1.5-72B |
|---|---|---|
| LMArena Math | 1218 | 1164 |
| OTIS Mock AIME 2024-2025 | 4.3% | — |
| MATH Level 5 | 22.6% | — |
Knowledge Llama 3-70B leads
Llama 3-70B: 20.8 (#272), Qwen1.5-72B: 11.5 (#295)
| Benchmark | Llama 3-70B | Qwen1.5-72B |
|---|---|---|
| GPQA Diamond | 40.6% | 28.8% |
| LMArena Expert | 1149 | 1136 |
| MMLU | 79.3% | — |
Multilingual Too close to call
Llama 3-70B: 33.8 (#246), Qwen1.5-72B: 33.4 (#248)
| Benchmark | Llama 3-70B | Qwen1.5-72B |
|---|---|---|
| LMArena Non-English | 1142 | 1135 |
| LMArena Chinese | 1114 | 1186 |
| LMArena French | 1231 | 1158 |
| LMArena German | 1169 | 1084 |
| LMArena Japanese | 1017 | 1061 |
| LMArena Korean | 1017 | 1049 |
| LMArena Russian | 1159 | 1104 |
| LMArena Spanish | 1241 | 1110 |
Instruction Following Llama 3-70B leads
Llama 3-70B: 62.8 (#233), Qwen1.5-72B: 59.6 (#251)
| Benchmark | Llama 3-70B | Qwen1.5-72B |
|---|---|---|
| LMArena Instruction Following | 1194 | 1142 |
Long Context Too close to call
Llama 3-70B: 35.6 (#235), Qwen1.5-72B: 35.0 (#238)
| Benchmark | Llama 3-70B | Qwen1.5-72B |
|---|---|---|
| LMArena Longer Query | 1174 | 1157 |
Writing & Preference Llama 3-70B leads
Llama 3-70B: 43.0 (#225), Qwen1.5-72B: 37.4 (#253)
| Benchmark | Llama 3-70B | Qwen1.5-72B |
|---|---|---|
| LMArena Text | 1221 | 1166 |
| LMArena Creative Writing | 1210 | 1137 |
| LMArena Multi-Turn | 1223 | 1160 |
Frequently asked questions
Is Llama 3-70B better than Qwen1.5-72B?
Qwen1.5-72B is the stronger model overall, scoring 30.3 to 28.5 on the Noometry Index.
Is Llama 3-70B or Qwen1.5-72B better for coding?
Llama 3-70B scores higher on coding benchmarks: 35.3 versus 31.4 in the Noometry coding category.
How many benchmarks do Llama 3-70B and Qwen1.5-72B share?
22 benchmarks have published results for both models. Llama 3-70B has 31 scored results on Noometry and Qwen1.5-72B has 22.