# phi-3-medium 14B vs Yi-Lightning

> Yi-Lightning is the stronger model overall, scoring 37.1 to 29.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/phi-3-medium-14b-vs-yi-lightning
- Last updated: 2026-10-12
- Shared benchmarks: 0

## Summary

- The widest gap is in knowledge, where Yi-Lightning leads 35.4 to 9.1.
- phi-3-medium 14B has downloadable open weights; the other is API-only.

## Snapshot

| | phi-3-medium 14B | Yi-Lightning |
|---|---|---|
| Provider | Microsoft | 01.AI |
| Noometry Index | 29.7 | 37.1 |
| Rank | 306 | 209 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Proprietary |

## Coding

- phi-3-medium 14B: 36.8 (#201)
- Yi-Lightning: 27.4 (#320)

| Benchmark | phi-3-medium 14B | Yi-Lightning |
|---|---|---|
| Aider Polyglot | — | 12.9% |
| BigCodeBench Instruct | 37.6% | — |
| LMArena Coding | — | 1312 |
| BigCodeBench Complete | 48.7% | — |

## Reasoning

- phi-3-medium 14B: —
- Yi-Lightning: 25.9 (#139)

| Benchmark | phi-3-medium 14B | Yi-Lightning |
|---|---|---|
| LMArena Hard Prompts | — | 1302 |
| 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: 27.3 (#250)
- Yi-Lightning: 36.2 (#172)

| Benchmark | phi-3-medium 14B | Yi-Lightning |
|---|---|---|
| LMArena Math | — | 1300 |
| MATH Level 5 | 17.6% | — |

## Knowledge

- phi-3-medium 14B: 9.1 (#306)
- Yi-Lightning: 35.4 (#185)

| Benchmark | phi-3-medium 14B | Yi-Lightning |
|---|---|---|
| GPQA Diamond | 27.6% | — |
| LMArena Expert | — | 1286 |
| ARC (AI2) Challenge | 91.6% | — |
| MMLU | 78% | — |
| OpenBookQA | 87.4% | — |
| TriviaQA | 73.9% | — |

## Multilingual

- phi-3-medium 14B: —
- Yi-Lightning: 42.2 (#196)

| Benchmark | phi-3-medium 14B | Yi-Lightning |
|---|---|---|
| LMArena Non-English | — | 1269 |
| LMArena Chinese | — | 1320 |
| LMArena French | — | 1305 |
| LMArena German | — | 1267 |
| LMArena Japanese | — | 1228 |
| LMArena Korean | — | 1192 |
| LMArena Russian | — | 1255 |
| LMArena Spanish | — | 1315 |

## Instruction Following

- phi-3-medium 14B: —
- Yi-Lightning: 67.4 (#196)

| Benchmark | phi-3-medium 14B | Yi-Lightning |
|---|---|---|
| LMArena Instruction Following | — | 1278 |

## Long Context

- phi-3-medium 14B: —
- Yi-Lightning: 39.4 (#181)

| Benchmark | phi-3-medium 14B | Yi-Lightning |
|---|---|---|
| LMArena Longer Query | — | 1297 |

## Writing & Preference

- phi-3-medium 14B: —
- Yi-Lightning: 50.2 (#184)

| Benchmark | phi-3-medium 14B | Yi-Lightning |
|---|---|---|
| LMArena Text | — | 1302 |
| LMArena Creative Writing | — | 1280 |
| LMArena Multi-Turn | — | 1311 |

## FAQ

### Is phi-3-medium 14B better than Yi-Lightning?

Yi-Lightning is the stronger model overall, scoring 37.1 to 29.7 on the Noometry Index.

### Is phi-3-medium 14B or Yi-Lightning better for coding?

phi-3-medium 14B scores higher on coding benchmarks: 36.8 versus 27.4 in the Noometry coding category.

### How many benchmarks do phi-3-medium 14B and Yi-Lightning share?

0 benchmarks have published results for both models. phi-3-medium 14B has 13 scored results on Noometry and Yi-Lightning has 18.