Model comparison
phi-3-medium 14B vs Qwen2.5 72B Instruct
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.5 to 29.1 on the Noometry Index.
Last verified . 11 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 11 benchmarks with published results for both. phi-3-medium 14B scores higher in 2 categories and Qwen2.5 72B Instruct in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5 72B Instruct leads 26.9 to 9.1.
- The biggest single-benchmark swing is MATH Level 5: 17.6% for phi-3-medium 14B and 63.2% for Qwen2.5 72B Instruct.
Side by side
| phi-3-medium 14B | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 29.1 | 31.5 |
| Released | 2024-04-23 | 2024-09 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $5.60 |
| Results tracked | 13 | 43 |
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), Qwen2.5 72B Instruct: 32.8 (#256)
| Benchmark | phi-3-medium 14B | Qwen2.5 72B Instruct |
|---|---|---|
| BigCodeBench Instruct | 37.6% | 45.8% |
| BigCodeBench Complete | 48.7% | 55.9% |
| WeirdML | — | 16% |
| LMArena Coding | — | 1293 |
Agentic & Tool Use Not comparable
phi-3-medium 14B: —, Qwen2.5 72B Instruct: 22.0 (#124)
| Benchmark | phi-3-medium 14B | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Not comparable
phi-3-medium 14B: —, Qwen2.5 72B Instruct: 21.4 (#197)
| Benchmark | phi-3-medium 14B | Qwen2.5 72B Instruct |
|---|---|---|
| BIG-Bench Hard | 81.4% | 79.8% |
| Epoch Capabilities Index | 121.18 | 129.01 |
| HellaSwag | 82.4% | 84.8% |
| WinoGrande | 81.5% | 82.3% |
| LMArena Hard Prompts | — | 1271 |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| Adversarial NLI | 55.8% | — |
| ForecastBench | — | 57.5 |
| PIQA | — | 82.6% |
Math phi-3-medium 14B leads
phi-3-medium 14B: 25.2 (#249), Qwen2.5 72B Instruct: 18.1 (#281)
| Benchmark | phi-3-medium 14B | Qwen2.5 72B Instruct |
|---|---|---|
| MATH Level 5 | 17.6% | 63.2% |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| LMArena Math | — | 1283 |
Knowledge Qwen2.5 72B Instruct leads
phi-3-medium 14B: 9.1 (#301), Qwen2.5 72B Instruct: 26.9 (#247)
| Benchmark | phi-3-medium 14B | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 27.6% | 49.1% |
| ARC (AI2) Challenge | 91.6% | 94.5% |
| MMLU | 78% | 85.3% |
| TriviaQA | 73.9% | 71.9% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| LMArena Expert | — | 1245 |
| OpenBookQA | 87.4% | — |
Multilingual Not comparable
phi-3-medium 14B: —, Qwen2.5 72B Instruct: 41.2 (#208)
| Benchmark | phi-3-medium 14B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | — | 1252 |
| LMArena Chinese | — | 1272 |
| LMArena French | — | 1279 |
| LMArena German | — | 1234 |
| LMArena Japanese | — | 1180 |
| LMArena Korean | — | 1188 |
| LMArena Russian | — | 1264 |
| LMArena Spanish | — | 1255 |
Instruction Following Not comparable
phi-3-medium 14B: —, Qwen2.5 72B Instruct: 65.7 (#216)
| Benchmark | phi-3-medium 14B | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | — | 80.6% |
| LMArena Instruction Following | — | 1254 |
Long Context Not comparable
phi-3-medium 14B: —, Qwen2.5 72B Instruct: 38.9 (#183)
| Benchmark | phi-3-medium 14B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | — | 1282 |
Writing & Preference Not comparable
phi-3-medium 14B: —, Qwen2.5 72B Instruct: 46.9 (#210)
| Benchmark | phi-3-medium 14B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | — | 1269 |
| LMArena Creative Writing | — | 1221 |
| WildBench | — | 80.2% |
| LMArena Multi-Turn | — | 1272 |
Frequently asked questions
Is phi-3-medium 14B better than Qwen2.5 72B Instruct?
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.5 to 29.1 on the Noometry Index.
Is phi-3-medium 14B or Qwen2.5 72B Instruct better for coding?
phi-3-medium 14B scores higher on coding benchmarks: 36.3 versus 32.8 in the Noometry coding category.
How many benchmarks do phi-3-medium 14B and Qwen2.5 72B Instruct share?
11 benchmarks have published results for both models. phi-3-medium 14B has 13 scored results on Noometry and Qwen2.5 72B Instruct has 43.