Model comparison
DeepSeek-V3.1 vs Step 3
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 40.5 on the Noometry Index.
Last verified . 16 shared benchmarks.
DeepSeek
StepFunSummary
- They share 16 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 6 categories and Step 3 in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.1 leads 43.7 to 36.8.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 53.2% for DeepSeek-V3.1 and 62.3% for Step 3.
Side by side
| DeepSeek-V3.1 | Step 3 | |
|---|---|---|
| Provider | DeepSeek | StepFun |
| Noometry Index | 42.8 | 40.5 |
| Released | 2025-08-21 | — |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.95 | — |
| Results tracked | 27 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek-V3.1: 40.3 (#144), Step 3: 40.1 (#147)
| Benchmark | DeepSeek-V3.1 | Step 3 |
|---|---|---|
| LMArena Coding | 1417 | 1367 |
| WeirdML | 38.4% | — |
Reasoning Too close to call
DeepSeek-V3.1: 27.9 (#110), Step 3: 28.4 (#105)
| Benchmark | DeepSeek-V3.1 | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 53.2% | 62.3% |
| LMArena Hard Prompts | 1417 | 1355 |
| SimpleBench | 40% | — |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
| Epoch Capabilities Index | 139.92 | — |
| ForecastBench | 58 | — |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 38.9 (#122), Step 3: 37.6 (#148)
| Benchmark | DeepSeek-V3.1 | Step 3 |
|---|---|---|
| LMArena Math | 1420 | 1366 |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Step 3: 36.8 (#164)
| Benchmark | DeepSeek-V3.1 | Step 3 |
|---|---|---|
| LMArena Expert | 1405 | 1333 |
| Vectara Hallucination Rate | 5.5% | — |
Multimodal Not comparable
DeepSeek-V3.1: —, Step 3: 35.5 (#86)
| Benchmark | DeepSeek-V3.1 | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), Step 3: 46.3 (#159)
| Benchmark | DeepSeek-V3.1 | Step 3 |
|---|---|---|
| LMArena Non-English | 1400 | 1327 |
| LMArena Chinese | 1469 | 1397 |
| LMArena German | 1411 | 1371 |
| LMArena Korean | 1337 | 1269 |
| LMArena Russian | 1405 | 1331 |
| LMArena Spanish | 1431 | 1371 |
| LMArena French | 1447 | — |
| LMArena Japanese | 1378 | — |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 73.9 (#110), Step 3: 70.4 (#164)
| Benchmark | DeepSeek-V3.1 | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1400 | 1332 |
Long Context Step 3 leads
DeepSeek-V3.1: 36.3 (#232), Step 3: 40.3 (#157)
| Benchmark | DeepSeek-V3.1 | Step 3 |
|---|---|---|
| LMArena Longer Query | 1422 | 1326 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), Step 3: 54.3 (#151)
| Benchmark | DeepSeek-V3.1 | Step 3 |
|---|---|---|
| LMArena Text | 1420 | 1350 |
| LMArena Creative Writing | 1401 | 1321 |
| LMArena Multi-Turn | 1408 | 1341 |
| EQ-Bench Creative Writing | 1436 | — |
Frequently asked questions
Is DeepSeek-V3.1 better than Step 3?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 40.5 on the Noometry Index.
Is DeepSeek-V3.1 or Step 3 better for coding?
They score almost the same on coding (40.3 vs 40.1); test both on your own repository before choosing.
How many benchmarks do DeepSeek-V3.1 and Step 3 share?
16 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Step 3 has 17.