# DeepSeek-V2.5 (Sep 2024) vs Yi-Lightning

> DeepSeek-V2.5 (Sep 2024) and Yi-Lightning score almost the same on the Noometry Index (37.6 vs 37.1), so choose on price, context window or the category you care about most.

- Canonical page: https://noometry.com/compare/deepseek-v2-5-vs-yi-lightning
- Last updated: 2026-10-11
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 4 categories and Yi-Lightning in 4 categories; one gap is clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-V2.5 (Sep 2024) leads 31.7 to 27.4.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek-V2.5 (Sep 2024) | Yi-Lightning |
|---|---|---|
| Provider | DeepSeek | 01.AI |
| Noometry Index | 37.6 | 37.1 |
| Rank | 200 | 209 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Proprietary |

## Coding

- DeepSeek-V2.5 (Sep 2024): 31.7 (#281)
- Yi-Lightning: 27.4 (#320)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Yi-Lightning |
|---|---|---|
| Aider Polyglot | 17.8% | 12.9% |
| LMArena Coding | 1309 | 1312 |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |

## Reasoning

- DeepSeek-V2.5 (Sep 2024): 25.6 (#145)
- Yi-Lightning: 25.9 (#139)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Yi-Lightning |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1302 |

## Math

- DeepSeek-V2.5 (Sep 2024): 35.9 (#177)
- Yi-Lightning: 36.2 (#172)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Yi-Lightning |
|---|---|---|
| LMArena Math | 1288 | 1300 |

## Knowledge

- DeepSeek-V2.5 (Sep 2024): 34.8 (#193)
- Yi-Lightning: 35.4 (#185)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Yi-Lightning |
|---|---|---|
| LMArena Expert | 1266 | 1286 |

## Multilingual

- DeepSeek-V2.5 (Sep 2024): 42.5 (#193)
- Yi-Lightning: 42.2 (#196)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Yi-Lightning |
|---|---|---|
| LMArena Non-English | 1273 | 1269 |
| LMArena Chinese | 1318 | 1320 |
| LMArena French | 1289 | 1305 |
| LMArena German | 1258 | 1267 |
| LMArena Japanese | 1228 | 1228 |
| LMArena Korean | 1209 | 1192 |
| LMArena Russian | 1289 | 1255 |
| LMArena Spanish | 1248 | 1315 |

## Instruction Following

- DeepSeek-V2.5 (Sep 2024): 67.5 (#194)
- Yi-Lightning: 67.4 (#196)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Yi-Lightning |
|---|---|---|
| LMArena Instruction Following | 1280 | 1278 |

## Long Context

- DeepSeek-V2.5 (Sep 2024): 39.5 (#174)
- Yi-Lightning: 39.4 (#181)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Yi-Lightning |
|---|---|---|
| LMArena Longer Query | 1301 | 1297 |

## Writing & Preference

- DeepSeek-V2.5 (Sep 2024): 49.8 (#187)
- Yi-Lightning: 50.2 (#184)

| Benchmark | DeepSeek-V2.5 (Sep 2024) | Yi-Lightning |
|---|---|---|
| LMArena Text | 1294 | 1302 |
| LMArena Creative Writing | 1285 | 1280 |
| LMArena Multi-Turn | 1297 | 1311 |

## FAQ

### Is DeepSeek-V2.5 (Sep 2024) better than Yi-Lightning?

DeepSeek-V2.5 (Sep 2024) and Yi-Lightning score almost the same on the Noometry Index (37.6 vs 37.1), so choose on price, context window or the category you care about most.

### Is DeepSeek-V2.5 (Sep 2024) or Yi-Lightning better for coding?

DeepSeek-V2.5 (Sep 2024) scores higher on coding benchmarks: 31.7 versus 27.4 in the Noometry coding category.

### How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Yi-Lightning share?

18 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Yi-Lightning has 18.