Model comparison
Step 3 vs Yi-Lightning
Step 3 is the stronger model overall, scoring 40.5 to 37.1 on the Noometry Index.
Last verified . 15 shared benchmarks.
StepFunSummary
- They share 15 benchmarks with published results for both. Step 3 scores higher in 8 categories and Yi-Lightning in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where Step 3 leads 40.1 to 27.4.
- Step 3 has downloadable open weights; the other is API-only.
Side by side
| Step 3 | Yi-Lightning | |
|---|---|---|
| Provider | StepFun | 01.AI |
| Noometry Index | 40.5 | 37.1 |
| Released | — | 2024-12-02 |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 17 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Step 3: 40.1 (#147), Yi-Lightning: 27.4 (#320)
| Benchmark | Step 3 | Yi-Lightning |
|---|---|---|
| LMArena Coding | 1367 | 1312 |
| Aider Polyglot | — | 12.9% |
Reasoning Step 3 leads
Step 3: 28.4 (#105), Yi-Lightning: 25.9 (#139)
| Benchmark | Step 3 | Yi-Lightning |
|---|---|---|
| LMArena Hard Prompts | 1355 | 1302 |
| Kagi LLM Benchmark | 62.3% | — |
Math Step 3 leads
Step 3: 37.6 (#148), Yi-Lightning: 36.2 (#172)
| Benchmark | Step 3 | Yi-Lightning |
|---|---|---|
| LMArena Math | 1366 | 1300 |
Knowledge Step 3 leads
Step 3: 36.8 (#164), Yi-Lightning: 35.4 (#185)
| Benchmark | Step 3 | Yi-Lightning |
|---|---|---|
| LMArena Expert | 1333 | 1286 |
Multimodal Not comparable
Step 3: 35.5 (#86), Yi-Lightning: —
| Benchmark | Step 3 | Yi-Lightning |
|---|---|---|
| LMArena Vision | 1177 | — |
Multilingual Step 3 leads
Step 3: 46.3 (#159), Yi-Lightning: 42.2 (#196)
| Benchmark | Step 3 | Yi-Lightning |
|---|---|---|
| LMArena Non-English | 1327 | 1269 |
| LMArena Chinese | 1397 | 1320 |
| LMArena German | 1371 | 1267 |
| LMArena Korean | 1269 | 1192 |
| LMArena Russian | 1331 | 1255 |
| LMArena Spanish | 1371 | 1315 |
| LMArena French | — | 1305 |
| LMArena Japanese | — | 1228 |
Instruction Following Step 3 leads
Step 3: 70.4 (#164), Yi-Lightning: 67.4 (#196)
| Benchmark | Step 3 | Yi-Lightning |
|---|---|---|
| LMArena Instruction Following | 1332 | 1278 |
Long Context Too close to call
Step 3: 40.3 (#157), Yi-Lightning: 39.4 (#181)
| Benchmark | Step 3 | Yi-Lightning |
|---|---|---|
| LMArena Longer Query | 1326 | 1297 |
Writing & Preference Step 3 leads
Step 3: 54.3 (#151), Yi-Lightning: 50.2 (#184)
| Benchmark | Step 3 | Yi-Lightning |
|---|---|---|
| LMArena Text | 1350 | 1302 |
| LMArena Creative Writing | 1321 | 1280 |
| LMArena Multi-Turn | 1341 | 1311 |
Frequently asked questions
Is Step 3 better than Yi-Lightning?
Step 3 is the stronger model overall, scoring 40.5 to 37.1 on the Noometry Index.
Is Step 3 or Yi-Lightning better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 27.4 in the Noometry coding category.
How many benchmarks do Step 3 and Yi-Lightning share?
15 benchmarks have published results for both models. Step 3 has 17 scored results on Noometry and Yi-Lightning has 18.