Model comparison
phi-3-medium 14B vs Qwen Max
Qwen Max is the stronger model overall, scoring 34.7 to 29.7 on the Noometry Index.
Last verified . 2 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 2 benchmarks with published results for both. phi-3-medium 14B scores higher in 2 categories and Qwen Max in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen Max leads 30.3 to 9.1.
- The biggest single-benchmark swing is MATH Level 5: 17.6% for phi-3-medium 14B and 67.2% for Qwen Max.
- phi-3-medium 14B has downloadable open weights; the other is API-only.
Side by side
| phi-3-medium 14B | Qwen Max | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 29.7 | 34.7 |
| Released | 2024-04-23 | 2024-04-03 |
| Weights | Open | Proprietary |
| Context window | — | 33K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.60 |
| Output $ / M tokens | — | $6.40 |
| Results tracked | 13 | 23 |
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: 36.8 (#201), Qwen Max: 30.7 (#292)
| Benchmark | phi-3-medium 14B | Qwen Max |
|---|---|---|
| Aider Polyglot | — | 21.8% |
| BigCodeBench Instruct | 37.6% | — |
| LMArena Coding | — | 1288 |
| BigCodeBench Complete | 48.7% | — |
Reasoning Not comparable
phi-3-medium 14B: —, Qwen Max: 25.1 (#151)
| Benchmark | phi-3-medium 14B | Qwen Max |
|---|---|---|
| LMArena Hard Prompts | — | 1269 |
| Adversarial NLI | 55.8% | — |
| BIG-Bench Hard | 81.4% | — |
| Epoch Capabilities Index | 121.23 | — |
| HellaSwag | 82.4% | — |
| WinoGrande | 81.5% | — |
Math phi-3-medium 14B leads
phi-3-medium 14B: 27.3 (#250), Qwen Max: 22.3 (#276)
| Benchmark | phi-3-medium 14B | Qwen Max |
|---|---|---|
| MATH Level 5 | 17.6% | 67.2% |
| OTIS Mock AIME 2024-2025 | — | 16.1% |
| LMArena Math | — | 1275 |
| FrontierMath (Feb 2025 set) | — | 1% |
Knowledge Qwen Max leads
phi-3-medium 14B: 9.1 (#306), Qwen Max: 30.3 (#228)
| Benchmark | phi-3-medium 14B | Qwen Max |
|---|---|---|
| GPQA Diamond | 27.6% | 56.1% |
| LMArena Expert | — | 1248 |
| ARC (AI2) Challenge | 91.6% | — |
| MMLU | 78% | — |
| OpenBookQA | 87.4% | — |
| TriviaQA | 73.9% | — |
Multilingual Not comparable
phi-3-medium 14B: —, Qwen Max: 41.8 (#202)
| Benchmark | phi-3-medium 14B | Qwen Max |
|---|---|---|
| LMArena Non-English | — | 1263 |
| LMArena Chinese | — | 1254 |
| LMArena French | — | 1330 |
| LMArena German | — | 1254 |
| LMArena Japanese | — | 1205 |
| LMArena Korean | — | 1142 |
| LMArena Russian | — | 1274 |
| LMArena Spanish | — | 1290 |
Instruction Following Not comparable
phi-3-medium 14B: —, Qwen Max: 66.5 (#208)
| Benchmark | phi-3-medium 14B | Qwen Max |
|---|---|---|
| LMArena Instruction Following | — | 1262 |
Long Context Not comparable
phi-3-medium 14B: —, Qwen Max: 39.4 (#180)
| Benchmark | phi-3-medium 14B | Qwen Max |
|---|---|---|
| Fiction.LiveBench | — | 66.7% |
| LMArena Longer Query | — | 1288 |
Writing & Preference Not comparable
phi-3-medium 14B: —, Qwen Max: 47.8 (#205)
| Benchmark | phi-3-medium 14B | Qwen Max |
|---|---|---|
| LMArena Text | — | 1282 |
| LMArena Creative Writing | — | 1248 |
| LMArena Multi-Turn | — | 1277 |
Frequently asked questions
Is phi-3-medium 14B better than Qwen Max?
Qwen Max is the stronger model overall, scoring 34.7 to 29.7 on the Noometry Index.
Is phi-3-medium 14B or Qwen Max better for coding?
phi-3-medium 14B scores higher on coding benchmarks: 36.8 versus 30.7 in the Noometry coding category.
How many benchmarks do phi-3-medium 14B and Qwen Max share?
2 benchmarks have published results for both models. phi-3-medium 14B has 13 scored results on Noometry and Qwen Max has 23.