Model comparison
Phi-2 vs Qwen-7B
Neither Phi-2 nor Qwen-7B has enough public benchmark results to be ranked yet; the rows below show what has been published.
Last verified . 4 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both.
Side by side
| Phi-2 | Qwen-7B | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | — | — |
| Released | 2023-12-12 | 2023-09-28 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 11 | 8 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Phi-2: —, Qwen-7B: —
| Benchmark | Phi-2 | Qwen-7B |
|---|---|---|
| HumanEval+ | 45.1% | — |
| MBPP+ | 54.2% | — |
Reasoning Not comparable
Phi-2: —, Qwen-7B: —
| Benchmark | Phi-2 | Qwen-7B |
|---|---|---|
| BIG-Bench Hard | 59.4% | 45% |
| Epoch Capabilities Index | 107.94 | 106.83 |
| Adversarial NLI | 42.5% | — |
| HellaSwag | 53.6% | — |
| LAMBADA | — | 67.9% |
| PIQA | — | 77.9% |
| WinoGrande | 54.7% | — |
Math Not comparable
Phi-2: —, Qwen-7B: —
| Benchmark | Phi-2 | Qwen-7B |
|---|---|---|
| GSM8K | — | 51.7% |
Knowledge Not comparable
Phi-2: —, Qwen-7B: —
| Benchmark | Phi-2 | Qwen-7B |
|---|---|---|
| ARC (AI2) Challenge | 75.9% | 75.3% |
| MMLU | 58.4% | 45% |
| BoolQ | — | 76.4% |
| OpenBookQA | 73.6% | — |
| TriviaQA | 45.2% | — |
Frequently asked questions
Is Phi-2 better than Qwen-7B?
Neither Phi-2 nor Qwen-7B has enough public benchmark results to be ranked yet; the rows below show what has been published.
How many benchmarks do Phi-2 and Qwen-7B share?
4 benchmarks have published results for both models. Phi-2 has 11 scored results on Noometry and Qwen-7B has 8.