Model comparison
Phi-2 vs Qwen2.5-Max
Qwen2.5-Max has enough public results to be ranked (#146); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 1 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 1 benchmark with published results for both.
- Phi-2 has downloadable open weights; the other is API-only.
Side by side
| Phi-2 | Qwen2.5-Max | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | — | 40.7 |
| Released | 2023-12-12 | 2025-01-25 |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 11 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Phi-2: —, Qwen2.5-Max: 41.8 (#117)
| Benchmark | Phi-2 | Qwen2.5-Max |
|---|---|---|
| LiveBench Coding | — | 64.4% |
| LMArena Coding | — | 1359 |
| HumanEval+ | 45.1% | — |
| MBPP+ | 54.2% | — |
Reasoning Not comparable
Phi-2: —, Qwen2.5-Max: 25.6 (#147)
| Benchmark | Phi-2 | Qwen2.5-Max |
|---|---|---|
| Epoch Capabilities Index | 107.94 | 132.53 |
| LiveBench Reasoning | — | 51.4% |
| LMArena Hard Prompts | — | 1360 |
| LiveBench Data Analysis | — | 67.9% |
| Adversarial NLI | 42.5% | — |
| BIG-Bench Hard | 59.4% | — |
| HellaSwag | 53.6% | — |
| LiveBench | — | 62.3% |
| WinoGrande | 54.7% | — |
Math Not comparable
Phi-2: —, Qwen2.5-Max: 36.9 (#162)
| Benchmark | Phi-2 | Qwen2.5-Max |
|---|---|---|
| LiveBench Math | — | 58.4% |
| LMArena Math | — | 1369 |
Knowledge Not comparable
Phi-2: —, Qwen2.5-Max: 35.3 (#186)
| Benchmark | Phi-2 | Qwen2.5-Max |
|---|---|---|
| Confabulations | — | 21.8% |
| LMArena Expert | — | 1337 |
| ARC (AI2) Challenge | 75.9% | — |
| MMLU | 58.4% | — |
| OpenBookQA | 73.6% | — |
| TriviaQA | 45.2% | — |
Multilingual Not comparable
Phi-2: —, Qwen2.5-Max: 48.1 (#146)
| Benchmark | Phi-2 | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | — | 1352 |
| LMArena Chinese | — | 1382 |
| LMArena French | — | 1396 |
| LMArena German | — | 1350 |
| LMArena Japanese | — | 1300 |
| LMArena Korean | — | 1304 |
| LMArena Russian | — | 1353 |
| LMArena Spanish | — | 1377 |
Instruction Following Not comparable
Phi-2: —, Qwen2.5-Max: 71.3 (#152)
| Benchmark | Phi-2 | Qwen2.5-Max |
|---|---|---|
| LiveBench Instruction Following | — | 75.3% |
| LMArena Instruction Following | — | 1335 |
Long Context Not comparable
Phi-2: —, Qwen2.5-Max: 41.4 (#142)
| Benchmark | Phi-2 | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | — | 1358 |
Writing & Preference Not comparable
Phi-2: —, Qwen2.5-Max: 55.4 (#146)
| Benchmark | Phi-2 | Qwen2.5-Max |
|---|---|---|
| LMArena Text | — | 1367 |
| LMArena Creative Writing | — | 1339 |
| Short-Story Creative Writing | — | 72.9% |
| LMArena Multi-Turn | — | 1364 |
| LiveBench Language | — | 56.3% |
Frequently asked questions
Is Phi-2 better than Qwen2.5-Max?
Qwen2.5-Max has enough public results to be ranked (#146); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do Phi-2 and Qwen2.5-Max share?
1 benchmark has published results for both models. Phi-2 has 11 scored results on Noometry and Qwen2.5-Max has 27.