Model comparison
Llama 2-7B vs Qwen1.5-72B
Qwen1.5-72B is the stronger model overall, scoring 30.8 to 29.1 on the Noometry Index.
Last verified . 15 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 15 benchmarks with published results for both. Llama 2-7B scores higher in 1 category and Qwen1.5-72B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama 2-7B leads 28.2 to 11.5.
Side by side
| Llama 2-7B | Qwen1.5-72B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 29.1 | 30.8 |
| Released | 2023-07-18 | 2024-02-04 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 29 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen1.5-72B leads
Llama 2-7B: 29.2 (#307), Qwen1.5-72B: 31.9 (#277)
| Benchmark | Llama 2-7B | Qwen1.5-72B |
|---|---|---|
| LMArena Coding | 1002 | 1165 |
| BigCodeBench Instruct | — | 33.2% |
| BigCodeBench Complete | — | 40.3% |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 61.6% |
Reasoning Qwen1.5-72B leads
Llama 2-7B: 15.7 (#312), Qwen1.5-72B: 22.2 (#203)
| Benchmark | Llama 2-7B | Qwen1.5-72B |
|---|---|---|
| LMArena Hard Prompts | 1009 | 1148 |
| Chess Puzzles | 0% | — |
| BIG-Bench Hard | 39.2% | — |
| Epoch Capabilities Index | 99.06 | — |
| HellaSwag | 77.2% | — |
| LAMBADA | 73.3% | — |
| PIQA | 78.8% | — |
| WinoGrande | 69.2% | — |
Math Qwen1.5-72B leads
Llama 2-7B: 30.7 (#233), Qwen1.5-72B: 33.2 (#205)
| Benchmark | Llama 2-7B | Qwen1.5-72B |
|---|---|---|
| LMArena Math | 1042 | 1164 |
| GSM8K | 16.7% | — |
Knowledge Llama 2-7B leads
Llama 2-7B: 28.2 (#248), Qwen1.5-72B: 11.5 (#300)
| Benchmark | Llama 2-7B | Qwen1.5-72B |
|---|---|---|
| LMArena Expert | 1036 | 1136 |
| GPQA Diamond | — | 28.8% |
| ARC (AI2) Challenge | 45.9% | — |
| BoolQ | 77.9% | — |
| MMLU | 45.8% | — |
| OpenBookQA | 58.6% | — |
| TriviaQA | 73.7% | — |
Multimodal Not comparable
Llama 2-7B: —, Qwen1.5-72B: —
| Benchmark | Llama 2-7B | Qwen1.5-72B |
|---|---|---|
| ScienceQA | 43.1% | — |
Multilingual Qwen1.5-72B leads
Llama 2-7B: 23.8 (#293), Qwen1.5-72B: 33.2 (#253)
| Benchmark | Llama 2-7B | Qwen1.5-72B |
|---|---|---|
| LMArena Non-English | 973 | 1135 |
| LMArena Chinese | 973 | 1186 |
| LMArena French | 970 | 1159 |
| LMArena German | 978 | 1084 |
| LMArena Russian | 995 | 1104 |
| LMArena Spanish | 1007 | 1110 |
| LMArena Japanese | — | 1061 |
| LMArena Korean | — | 1050 |
Instruction Following Qwen1.5-72B leads
Llama 2-7B: 50.8 (#298), Qwen1.5-72B: 59.3 (#256)
| Benchmark | Llama 2-7B | Qwen1.5-72B |
|---|---|---|
| LMArena Instruction Following | 1006 | 1141 |
Long Context Qwen1.5-72B leads
Llama 2-7B: 30.4 (#287), Qwen1.5-72B: 35.1 (#243)
| Benchmark | Llama 2-7B | Qwen1.5-72B |
|---|---|---|
| LMArena Longer Query | 999 | 1157 |
Writing & Preference Qwen1.5-72B leads
Llama 2-7B: 28.0 (#298), Qwen1.5-72B: 37.3 (#258)
| Benchmark | Llama 2-7B | Qwen1.5-72B |
|---|---|---|
| LMArena Text | 1053 | 1166 |
| LMArena Creative Writing | 1033 | 1137 |
| LMArena Multi-Turn | 1029 | 1160 |
Frequently asked questions
Is Llama 2-7B better than Qwen1.5-72B?
Qwen1.5-72B is the stronger model overall, scoring 30.8 to 29.1 on the Noometry Index.
Is Llama 2-7B or Qwen1.5-72B better for coding?
Qwen1.5-72B scores higher on coding benchmarks: 31.9 versus 29.2 in the Noometry coding category.
How many benchmarks do Llama 2-7B and Qwen1.5-72B share?
15 benchmarks have published results for both models. Llama 2-7B has 29 scored results on Noometry and Qwen1.5-72B has 22.