Model comparison
Phi-2 vs Qwen1.5-72B
Qwen1.5-72B has enough public results to be ranked (#282); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 2 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 2 benchmarks with published results for both.
Side by side
| Phi-2 | Qwen1.5-72B | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | — | 30.3 |
| Released | 2023-12-12 | 2024-02-04 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 11 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Phi-2: —, Qwen1.5-72B: 31.4 (#272)
| Benchmark | Phi-2 | Qwen1.5-72B |
|---|---|---|
| HumanEval+ | 45.1% | 59.1% |
| MBPP+ | 54.2% | 61.6% |
| BigCodeBench Instruct | — | 33.2% |
| LMArena Coding | — | 1166 |
| BigCodeBench Complete | — | 40.3% |
Reasoning Not comparable
Phi-2: —, Qwen1.5-72B: 21.2 (#201)
| Benchmark | Phi-2 | Qwen1.5-72B |
|---|---|---|
| LMArena Hard Prompts | — | 1148 |
| Adversarial NLI | 42.5% | — |
| BIG-Bench Hard | 59.4% | — |
| Epoch Capabilities Index | 107.62 | — |
| HellaSwag | 53.6% | — |
| WinoGrande | 54.7% | — |
Math Not comparable
Phi-2: —, Qwen1.5-72B: 30.7 (#205)
| Benchmark | Phi-2 | Qwen1.5-72B |
|---|---|---|
| LMArena Math | — | 1164 |
Knowledge Not comparable
Phi-2: —, Qwen1.5-72B: 11.5 (#295)
| Benchmark | Phi-2 | Qwen1.5-72B |
|---|---|---|
| GPQA Diamond | — | 28.8% |
| LMArena Expert | — | 1136 |
| ARC (AI2) Challenge | 75.9% | — |
| MMLU | 58.4% | — |
| OpenBookQA | 73.6% | — |
| TriviaQA | 45.2% | — |
Multilingual Not comparable
Phi-2: —, Qwen1.5-72B: 33.4 (#248)
| Benchmark | Phi-2 | Qwen1.5-72B |
|---|---|---|
| LMArena Non-English | — | 1135 |
| LMArena Chinese | — | 1186 |
| LMArena French | — | 1158 |
| LMArena German | — | 1084 |
| LMArena Japanese | — | 1061 |
| LMArena Korean | — | 1049 |
| LMArena Russian | — | 1104 |
| LMArena Spanish | — | 1110 |
Instruction Following Not comparable
Phi-2: —, Qwen1.5-72B: 59.6 (#251)
| Benchmark | Phi-2 | Qwen1.5-72B |
|---|---|---|
| LMArena Instruction Following | — | 1142 |
Long Context Not comparable
Phi-2: —, Qwen1.5-72B: 35.0 (#238)
| Benchmark | Phi-2 | Qwen1.5-72B |
|---|---|---|
| LMArena Longer Query | — | 1157 |
Writing & Preference Not comparable
Phi-2: —, Qwen1.5-72B: 37.4 (#253)
| Benchmark | Phi-2 | Qwen1.5-72B |
|---|---|---|
| LMArena Text | — | 1166 |
| LMArena Creative Writing | — | 1137 |
| LMArena Multi-Turn | — | 1160 |
Frequently asked questions
Is Phi-2 better than Qwen1.5-72B?
Qwen1.5-72B has enough public results to be ranked (#282); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do Phi-2 and Qwen1.5-72B share?
2 benchmarks have published results for both models. Phi-2 has 11 scored results on Noometry and Qwen1.5-72B has 22.