Model comparison
Phi-2 vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B has enough public results to be ranked (#245); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 7 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 7 benchmarks with published results for both.
Side by side
| Phi-2 | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | — | 33.4 |
| Released | 2023-12-12 | 2024-09-18 |
| Weights | Open | Open |
| Context window | — | 33K |
| Max output | — | 29K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1 |
| Results tracked | 11 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Phi-2: —, Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Phi-2 | Qwen2.5-Coder-32B |
|---|---|---|
| HumanEval+ | 45.1% | 87.2% |
| MBPP+ | 54.2% | 77% |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| LMArena Coding | — | 1276 |
| BigCodeBench Complete | — | 58% |
Reasoning Not comparable
Phi-2: —, Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Phi-2 | Qwen2.5-Coder-32B |
|---|---|---|
| Epoch Capabilities Index | 107.94 | 119.49 |
| HellaSwag | 53.6% | 83% |
| WinoGrande | 54.7% | 80.8% |
| LiveBench Reasoning | — | 42.1% |
| LMArena Hard Prompts | — | 1251 |
| LiveBench Data Analysis | — | 49.9% |
| Adversarial NLI | 42.5% | — |
| BIG-Bench Hard | 59.4% | — |
| LiveBench | — | 46.2% |
Math Not comparable
Phi-2: —, Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Phi-2 | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Math | — | 46.6% |
| LMArena Math | — | 1251 |
| GSM8K | — | 93% |
Knowledge Not comparable
Phi-2: —, Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Phi-2 | Qwen2.5-Coder-32B |
|---|---|---|
| ARC (AI2) Challenge | 75.9% | 70.5% |
| MMLU | 58.4% | 79.1% |
| LMArena Expert | — | 1221 |
| OpenBookQA | 73.6% | — |
| TriviaQA | 45.2% | — |
Multilingual Not comparable
Phi-2: —, Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Phi-2 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | — | 1205 |
| LMArena Chinese | — | 1222 |
| LMArena Russian | — | 1228 |
Instruction Following Not comparable
Phi-2: —, Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Phi-2 | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Instruction Following | — | 58.7% |
| LMArena Instruction Following | — | 1223 |
Long Context Not comparable
Phi-2: —, Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Phi-2 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | — | 1251 |
Writing & Preference Not comparable
Phi-2: —, Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Phi-2 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | — | 1230 |
| LMArena Creative Writing | — | 1174 |
| LMArena Multi-Turn | — | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Phi-2 better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B has enough public results to be ranked (#245); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do Phi-2 and Qwen2.5-Coder-32B share?
7 benchmarks have published results for both models. Phi-2 has 11 scored results on Noometry and Qwen2.5-Coder-32B has 31.