Model comparison
DeepSeek-V3.1 vs Yi-Lightning
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 37.1 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeekSummary
- They share 17 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 7 categories and Yi-Lightning in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-V3.1 leads 40.3 to 27.4.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1 | Yi-Lightning | |
|---|---|---|
| Provider | DeepSeek | 01.AI |
| Noometry Index | 42.8 | 37.1 |
| Released | 2025-08-21 | 2024-12-02 |
| Weights | Open | Proprietary |
| Context window | 164K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.95 | — |
| Results tracked | 27 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 40.3 (#144), Yi-Lightning: 27.4 (#320)
| Benchmark | DeepSeek-V3.1 | Yi-Lightning |
|---|---|---|
| LMArena Coding | 1417 | 1312 |
| Aider Polyglot | — | 12.9% |
| WeirdML | 38.4% | — |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), Yi-Lightning: 25.9 (#139)
| Benchmark | DeepSeek-V3.1 | Yi-Lightning |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1302 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
| Epoch Capabilities Index | 139.92 | — |
| ForecastBench | 58 | — |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 38.9 (#122), Yi-Lightning: 36.2 (#172)
| Benchmark | DeepSeek-V3.1 | Yi-Lightning |
|---|---|---|
| LMArena Math | 1420 | 1300 |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Yi-Lightning: 35.4 (#185)
| Benchmark | DeepSeek-V3.1 | Yi-Lightning |
|---|---|---|
| LMArena Expert | 1405 | 1286 |
| Vectara Hallucination Rate | 5.5% | — |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), Yi-Lightning: 42.2 (#196)
| Benchmark | DeepSeek-V3.1 | Yi-Lightning |
|---|---|---|
| LMArena Non-English | 1400 | 1269 |
| LMArena Chinese | 1469 | 1320 |
| LMArena French | 1447 | 1305 |
| LMArena German | 1411 | 1267 |
| LMArena Japanese | 1378 | 1228 |
| LMArena Korean | 1337 | 1192 |
| LMArena Russian | 1405 | 1255 |
| LMArena Spanish | 1431 | 1315 |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 73.9 (#110), Yi-Lightning: 67.4 (#196)
| Benchmark | DeepSeek-V3.1 | Yi-Lightning |
|---|---|---|
| LMArena Instruction Following | 1400 | 1278 |
Long Context Yi-Lightning leads
DeepSeek-V3.1: 36.3 (#232), Yi-Lightning: 39.4 (#181)
| Benchmark | DeepSeek-V3.1 | Yi-Lightning |
|---|---|---|
| LMArena Longer Query | 1422 | 1297 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), Yi-Lightning: 50.2 (#184)
| Benchmark | DeepSeek-V3.1 | Yi-Lightning |
|---|---|---|
| LMArena Text | 1420 | 1302 |
| LMArena Creative Writing | 1401 | 1280 |
| LMArena Multi-Turn | 1408 | 1311 |
| EQ-Bench Creative Writing | 1436 | — |
Frequently asked questions
Is DeepSeek-V3.1 better than Yi-Lightning?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 37.1 on the Noometry Index.
Is DeepSeek-V3.1 or Yi-Lightning better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 27.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1 and Yi-Lightning share?
17 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Yi-Lightning has 18.