Model comparison
phi-3-medium 14B vs Qwen2.5 14B Instruct
phi-3-medium 14B has enough public results to be ranked (#306); Qwen2.5 14B Instruct does not yet, so treat this comparison as directional.
Last verified . 3 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 3 benchmarks with published results for both. phi-3-medium 14B scores higher in 0 categories and Qwen2.5 14B Instruct in 1 category, but none of those gaps is larger than the uncertainty.
Side by side
| phi-3-medium 14B | Qwen2.5 14B Instruct | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 29.7 | 38.7 |
| Released | 2024-04-23 | 2024-09 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $0.35 |
| Output $ / M tokens | — | $1.40 |
| Results tracked | 13 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
phi-3-medium 14B: 36.8 (#201), Qwen2.5 14B Instruct: 37.7 (#191)
| Benchmark | phi-3-medium 14B | Qwen2.5 14B Instruct |
|---|---|---|
| BigCodeBench Instruct | 37.6% | 39.8% |
| BigCodeBench Complete | 48.7% | 52.2% |
Reasoning Not comparable
phi-3-medium 14B: —, Qwen2.5 14B Instruct: —
| Benchmark | phi-3-medium 14B | Qwen2.5 14B Instruct |
|---|---|---|
| Adversarial NLI | 55.8% | — |
| BIG-Bench Hard | 81.4% | — |
| Epoch Capabilities Index | 121.23 | — |
| HellaSwag | 82.4% | — |
| WinoGrande | 81.5% | — |
Math Not comparable
phi-3-medium 14B: 27.3 (#250), Qwen2.5 14B Instruct: —
| Benchmark | phi-3-medium 14B | Qwen2.5 14B Instruct |
|---|---|---|
| MATH Level 5 | 17.6% | — |
Knowledge Not comparable
phi-3-medium 14B: 9.1 (#306), Qwen2.5 14B Instruct: —
| Benchmark | phi-3-medium 14B | Qwen2.5 14B Instruct |
|---|---|---|
| MMLU | 78% | 79.9% |
| GPQA Diamond | 27.6% | — |
| ARC (AI2) Challenge | 91.6% | — |
| OpenBookQA | 87.4% | — |
| TriviaQA | 73.9% | — |
Frequently asked questions
Is phi-3-medium 14B better than Qwen2.5 14B Instruct?
phi-3-medium 14B has enough public results to be ranked (#306); Qwen2.5 14B Instruct does not yet, so treat this comparison as directional.
Is phi-3-medium 14B or Qwen2.5 14B Instruct better for coding?
They score almost the same on coding (36.8 vs 37.7); test both on your own repository before choosing.
How many benchmarks do phi-3-medium 14B and Qwen2.5 14B Instruct share?
3 benchmarks have published results for both models. phi-3-medium 14B has 13 scored results on Noometry and Qwen2.5 14B Instruct has 3.