Model comparison
Phi-2 vs Qwen-14B
Qwen-14B has enough public results to be ranked (#275); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 4 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both.
Side by side
| Phi-2 | Qwen-14B | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | — | 31.4 |
| Released | 2023-12-12 | 2023-09-24 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 11 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Phi-2: —, Qwen-14B: 31.2 (#288)
| Benchmark | Phi-2 | Qwen-14B |
|---|---|---|
| LMArena Coding | — | 1071 |
| HumanEval+ | 45.1% | — |
| MBPP+ | 54.2% | — |
Reasoning Not comparable
Phi-2: —, Qwen-14B: 19.6 (#257)
| Benchmark | Phi-2 | Qwen-14B |
|---|---|---|
| BIG-Bench Hard | 59.4% | 55% |
| Epoch Capabilities Index | 107.94 | 113.03 |
| LMArena Hard Prompts | — | 1027 |
| Adversarial NLI | 42.5% | — |
| HellaSwag | 53.6% | — |
| LAMBADA | — | 71.1% |
| PIQA | — | 79.9% |
| WinoGrande | 54.7% | — |
Math Not comparable
Phi-2: —, Qwen-14B: 31.2 (#227)
| Benchmark | Phi-2 | Qwen-14B |
|---|---|---|
| LMArena Math | — | 1068 |
| GSM8K | — | 61.3% |
Knowledge Not comparable
Phi-2: —, Qwen-14B: —
| Benchmark | Phi-2 | Qwen-14B |
|---|---|---|
| ARC (AI2) Challenge | 75.9% | 84.4% |
| MMLU | 58.4% | 66.3% |
| BoolQ | — | 86.2% |
| OpenBookQA | 73.6% | — |
| TriviaQA | 45.2% | — |
Multilingual Not comparable
Phi-2: —, Qwen-14B: 27.5 (#275)
| Benchmark | Phi-2 | Qwen-14B |
|---|---|---|
| LMArena Non-English | — | 1041 |
| LMArena Chinese | — | 1077 |
Instruction Following Not comparable
Phi-2: —, Qwen-14B: 52.4 (#289)
| Benchmark | Phi-2 | Qwen-14B |
|---|---|---|
| LMArena Instruction Following | — | 1031 |
Long Context Not comparable
Phi-2: —, Qwen-14B: 31.3 (#280)
| Benchmark | Phi-2 | Qwen-14B |
|---|---|---|
| LMArena Longer Query | — | 1028 |
Writing & Preference Not comparable
Phi-2: —, Qwen-14B: 27.6 (#299)
| Benchmark | Phi-2 | Qwen-14B |
|---|---|---|
| LMArena Text | — | 1051 |
| LMArena Creative Writing | — | 1028 |
| LMArena Multi-Turn | — | 1022 |
Frequently asked questions
Is Phi-2 better than Qwen-14B?
Qwen-14B has enough public results to be ranked (#275); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do Phi-2 and Qwen-14B share?
4 benchmarks have published results for both models. Phi-2 has 11 scored results on Noometry and Qwen-14B has 18.