Model comparison
phi-3-medium 14B vs Qwen3.8 27B
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 29.7 on the Noometry Index.
Last verified . 1 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 1 benchmark with published results for both. phi-3-medium 14B scores higher in 0 categories and Qwen3.8 27B in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.8 27B leads 41.6 to 9.1.
Side by side
| phi-3-medium 14B | Qwen3.8 27B | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 29.7 | 46.0 |
| Released | 2024-04-23 | 2026-08-14 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 33K |
| Input $ / M tokens | — | $0.99 |
| Output $ / M tokens | — | $1.49 |
| Results tracked | 13 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
phi-3-medium 14B: 36.8 (#201), Qwen3.8 27B: 50.5 (#44)
| Benchmark | phi-3-medium 14B | Qwen3.8 27B |
|---|---|---|
| LMArena WebDev | — | 1593 |
| SciCode | — | 46.6% |
| BigCodeBench Instruct | 37.6% | — |
| LMArena Coding | — | 1482 |
| BigCodeBench Complete | 48.7% | — |
Agentic & Tool Use Not comparable
phi-3-medium 14B: —, Qwen3.8 27B: 32.9 (#57)
| Benchmark | phi-3-medium 14B | Qwen3.8 27B |
|---|---|---|
| APEX-Agents | — | 47.5% |
Reasoning Not comparable
phi-3-medium 14B: —, Qwen3.8 27B: 41.0 (#54)
| Benchmark | phi-3-medium 14B | Qwen3.8 27B |
|---|---|---|
| Epoch Capabilities Index | 121.23 | 149.38 |
| ARC-AGI-2 | — | 42.4% |
| NYT Connections (extended) | — | 54.5% |
| ARC-AGI-1 | — | 87.5% |
| CritPt | — | 5.4% |
| LMArena Hard Prompts | — | 1460 |
| DTBench | — | 88% |
| LMCA | — | 41.4% |
| Surface Evolver Bench | — | 45% |
| Adversarial NLI | 55.8% | — |
| BIG-Bench Hard | 81.4% | — |
| HellaSwag | 82.4% | — |
| WinoGrande | 81.5% | — |
Math Qwen3.8 27B leads
phi-3-medium 14B: 27.3 (#250), Qwen3.8 27B: 37.1 (#161)
| Benchmark | phi-3-medium 14B | Qwen3.8 27B |
|---|---|---|
| ProofBench | — | 16% |
| LMArena Math | — | 1456 |
| MATH Level 5 | 17.6% | — |
Knowledge Qwen3.8 27B leads
phi-3-medium 14B: 9.1 (#306), Qwen3.8 27B: 41.6 (#109)
| Benchmark | phi-3-medium 14B | Qwen3.8 27B |
|---|---|---|
| GPQA Diamond | 27.6% | — |
| LMArena Expert | — | 1482 |
| ARC (AI2) Challenge | 91.6% | — |
| MMLU | 78% | — |
| OpenBookQA | 87.4% | — |
| TriviaQA | 73.9% | — |
Multimodal Not comparable
phi-3-medium 14B: —, Qwen3.8 27B: 41.3 (#37)
| Benchmark | phi-3-medium 14B | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | — | 1271 |
Multilingual Not comparable
phi-3-medium 14B: —, Qwen3.8 27B: 53.7 (#60)
| Benchmark | phi-3-medium 14B | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | — | 1430 |
| LMArena Chinese | — | 1504 |
| LMArena French | — | 1465 |
| LMArena German | — | 1438 |
| LMArena Japanese | — | 1384 |
| LMArena Korean | — | 1393 |
| LMArena Russian | — | 1415 |
| LMArena Spanish | — | 1448 |
Instruction Following Not comparable
phi-3-medium 14B: —, Qwen3.8 27B: 75.8 (#53)
| Benchmark | phi-3-medium 14B | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | — | 1439 |
Long Context Not comparable
phi-3-medium 14B: —, Qwen3.8 27B: 44.3 (#70)
| Benchmark | phi-3-medium 14B | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | — | 1450 |
Writing & Preference Not comparable
phi-3-medium 14B: —, Qwen3.8 27B: 65.8 (#43)
| Benchmark | phi-3-medium 14B | Qwen3.8 27B |
|---|---|---|
| LMArena Text | — | 1441 |
| LMArena Creative Writing | — | 1384 |
| EQ-Bench Creative Writing | — | 1671 |
| LMArena Multi-Turn | — | 1441 |
Frequently asked questions
Is phi-3-medium 14B better than Qwen3.8 27B?
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 29.7 on the Noometry Index.
Is phi-3-medium 14B or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 36.8 in the Noometry coding category.
How many benchmarks do phi-3-medium 14B and Qwen3.8 27B share?
1 benchmark has published results for both models. phi-3-medium 14B has 13 scored results on Noometry and Qwen3.8 27B has 31.