Model comparison
phi-3-medium 14B vs Qwen-14B
Qwen-14B is the stronger model overall, scoring 30.8 to 29.1 on the Noometry Index.
Last verified . 4 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both. phi-3-medium 14B scores higher in 1 category and Qwen-14B in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where phi-3-medium 14B leads 36.3 to 30.6.
Side by side
| phi-3-medium 14B | Qwen-14B | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 29.1 | 30.8 |
| Released | 2024-04-23 | 2023-09-24 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 13 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding phi-3-medium 14B leads
phi-3-medium 14B: 36.3 (#198), Qwen-14B: 30.6 (#283)
| Benchmark | phi-3-medium 14B | Qwen-14B |
|---|---|---|
| BigCodeBench Instruct | 37.6% | — |
| LMArena Coding | — | 1072 |
| BigCodeBench Complete | 48.7% | — |
Reasoning Not comparable
phi-3-medium 14B: —, Qwen-14B: 18.7 (#255)
| Benchmark | phi-3-medium 14B | Qwen-14B |
|---|---|---|
| BIG-Bench Hard | 81.4% | 55% |
| Epoch Capabilities Index | 121.18 | 112.8 |
| LMArena Hard Prompts | — | 1027 |
| Adversarial NLI | 55.8% | — |
| HellaSwag | 82.4% | — |
| LAMBADA | — | 71.1% |
| PIQA | — | 79.9% |
| WinoGrande | 81.5% | — |
Math Qwen-14B leads
phi-3-medium 14B: 25.2 (#249), Qwen-14B: 28.7 (#224)
| Benchmark | phi-3-medium 14B | Qwen-14B |
|---|---|---|
| LMArena Math | — | 1068 |
| MATH Level 5 | 17.6% | — |
| GSM8K | — | 61.3% |
Knowledge Not comparable
phi-3-medium 14B: 9.1 (#301), Qwen-14B: —
| Benchmark | phi-3-medium 14B | Qwen-14B |
|---|---|---|
| ARC (AI2) Challenge | 91.6% | 84.4% |
| MMLU | 78% | 66.3% |
| GPQA Diamond | 27.6% | — |
| BoolQ | — | 86.2% |
| OpenBookQA | 87.4% | — |
| TriviaQA | 73.9% | — |
Multilingual Not comparable
phi-3-medium 14B: —, Qwen-14B: 27.7 (#270)
| Benchmark | phi-3-medium 14B | Qwen-14B |
|---|---|---|
| LMArena Non-English | — | 1042 |
| LMArena Chinese | — | 1077 |
Instruction Following Not comparable
phi-3-medium 14B: —, Qwen-14B: 52.6 (#284)
| Benchmark | phi-3-medium 14B | Qwen-14B |
|---|---|---|
| LMArena Instruction Following | — | 1031 |
Long Context Not comparable
phi-3-medium 14B: —, Qwen-14B: 31.3 (#276)
| Benchmark | phi-3-medium 14B | Qwen-14B |
|---|---|---|
| LMArena Longer Query | — | 1029 |
Writing & Preference Not comparable
phi-3-medium 14B: —, Qwen-14B: 27.8 (#294)
| Benchmark | phi-3-medium 14B | Qwen-14B |
|---|---|---|
| LMArena Text | — | 1051 |
| LMArena Creative Writing | — | 1028 |
| LMArena Multi-Turn | — | 1022 |
Frequently asked questions
Is phi-3-medium 14B better than Qwen-14B?
Qwen-14B is the stronger model overall, scoring 30.8 to 29.1 on the Noometry Index.
Is phi-3-medium 14B or Qwen-14B better for coding?
phi-3-medium 14B scores higher on coding benchmarks: 36.3 versus 30.6 in the Noometry coding category.
How many benchmarks do phi-3-medium 14B and Qwen-14B share?
4 benchmarks have published results for both models. phi-3-medium 14B has 13 scored results on Noometry and Qwen-14B has 18.