Model comparison
Tulu 3 (Tülu 3) 70B vs Yi-Lightning
Yi-Lightning is the stronger model overall, scoring 37.1 to 33.0 on the Noometry Index.
Last verified . 11 shared benchmarks.
Allen Institute for AI (Ai2)Summary
- They share 11 benchmarks with published results for both. Tulu 3 (Tülu 3) 70B scores higher in 1 category and Yi-Lightning in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Yi-Lightning leads 36.2 to 14.2.
- Tulu 3 (Tülu 3) 70B has downloadable open weights; the other is API-only.
Side by side
| Tulu 3 (Tülu 3) 70B | Yi-Lightning | |
|---|---|---|
| Provider | Allen Institute for AI (Ai2) | 01.AI |
| Noometry Index | 33.0 | 37.1 |
| Released | 2024-11-21 | 2024-12-02 |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 14 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Tulu 3 (Tülu 3) 70B leads
Tulu 3 (Tülu 3) 70B: 36.0 (#214), Yi-Lightning: 27.4 (#320)
| Benchmark | Tulu 3 (Tülu 3) 70B | Yi-Lightning |
|---|---|---|
| LMArena Coding | 1235 | 1312 |
| Aider Polyglot | — | 12.9% |
Reasoning Yi-Lightning leads
Tulu 3 (Tülu 3) 70B: 23.9 (#169), Yi-Lightning: 25.9 (#139)
| Benchmark | Tulu 3 (Tülu 3) 70B | Yi-Lightning |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1302 |
Math Yi-Lightning leads
Tulu 3 (Tülu 3) 70B: 14.2 (#303), Yi-Lightning: 36.2 (#172)
| Benchmark | Tulu 3 (Tülu 3) 70B | Yi-Lightning |
|---|---|---|
| LMArena Math | 1242 | 1300 |
| OTIS Mock AIME 2024-2025 | 4.4% | — |
| MATH Level 5 | 42.7% | — |
Knowledge Yi-Lightning leads
Tulu 3 (Tülu 3) 70B: 25.0 (#264), Yi-Lightning: 35.4 (#185)
| Benchmark | Tulu 3 (Tülu 3) 70B | Yi-Lightning |
|---|---|---|
| GPQA Diamond | 46.3% | — |
| LMArena Expert | — | 1286 |
Multilingual Yi-Lightning leads
Tulu 3 (Tülu 3) 70B: 39.9 (#222), Yi-Lightning: 42.2 (#196)
| Benchmark | Tulu 3 (Tülu 3) 70B | Yi-Lightning |
|---|---|---|
| LMArena Non-English | 1236 | 1269 |
| LMArena Chinese | 1249 | 1320 |
| LMArena Russian | 1246 | 1255 |
| LMArena French | — | 1305 |
| LMArena German | — | 1267 |
| LMArena Japanese | — | 1228 |
| LMArena Korean | — | 1192 |
| LMArena Spanish | — | 1315 |
Instruction Following Yi-Lightning leads
Tulu 3 (Tülu 3) 70B: 64.8 (#227), Yi-Lightning: 67.4 (#196)
| Benchmark | Tulu 3 (Tülu 3) 70B | Yi-Lightning |
|---|---|---|
| LMArena Instruction Following | 1233 | 1278 |
Long Context Yi-Lightning leads
Tulu 3 (Tülu 3) 70B: 37.1 (#222), Yi-Lightning: 39.4 (#181)
| Benchmark | Tulu 3 (Tülu 3) 70B | Yi-Lightning |
|---|---|---|
| LMArena Longer Query | 1224 | 1297 |
Writing & Preference Yi-Lightning leads
Tulu 3 (Tülu 3) 70B: 45.6 (#223), Yi-Lightning: 50.2 (#184)
| Benchmark | Tulu 3 (Tülu 3) 70B | Yi-Lightning |
|---|---|---|
| LMArena Text | 1256 | 1302 |
| LMArena Creative Writing | 1231 | 1280 |
| LMArena Multi-Turn | 1252 | 1311 |
Frequently asked questions
Is Tulu 3 (Tülu 3) 70B better than Yi-Lightning?
Yi-Lightning is the stronger model overall, scoring 37.1 to 33.0 on the Noometry Index.
Is Tulu 3 (Tülu 3) 70B or Yi-Lightning better for coding?
Tulu 3 (Tülu 3) 70B scores higher on coding benchmarks: 36.0 versus 27.4 in the Noometry coding category.
How many benchmarks do Tulu 3 (Tülu 3) 70B and Yi-Lightning share?
11 benchmarks have published results for both models. Tulu 3 (Tülu 3) 70B has 14 scored results on Noometry and Yi-Lightning has 18.