Model comparison
Deepseek Coder v2 vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 35.9 on the Noometry Index.
Last verified . 13 shared benchmarks.
DeepSeek
StepFunSummary
- They share 13 benchmarks with published results for both. Deepseek Coder v2 scores higher in 0 categories and Step 5 Preview in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Step 5 Preview leads 62.8 to 38.2.
- Deepseek Coder v2 has downloadable open weights; the other is API-only.
Side by side
| Deepseek Coder v2 | Step 5 Preview | |
|---|---|---|
| Provider | DeepSeek | StepFun |
| Noometry Index | 35.9 | 47.9 |
| Released | 2024-06-17 | 2026-09-16 |
| Weights | Open | Proprietary |
| Context window | — | 1.02M |
| Max output | — | 66K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $2.70 |
| Results tracked | 24 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
Deepseek Coder v2: 38.1 (#183), Step 5 Preview: 51.8 (#36)
| Benchmark | Deepseek Coder v2 | Step 5 Preview |
|---|---|---|
| LMArena Coding | 1251 | 1480 |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning Step 5 Preview leads
Deepseek Coder v2: 23.6 (#176), Step 5 Preview: 40.0 (#57)
| Benchmark | Deepseek Coder v2 | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1465 |
| CritPt | — | 20.9% |
| WinoGrande | 83.7% | — |
Math Step 5 Preview leads
Deepseek Coder v2: 34.9 (#190), Step 5 Preview: 46.1 (#72)
| Benchmark | Deepseek Coder v2 | Step 5 Preview |
|---|---|---|
| LMArena Math | 1241 | 1470 |
| ProofBench | — | 42% |
| GSM8K | 94.5% | — |
Knowledge Step 5 Preview leads
Deepseek Coder v2: 32.3 (#212), Step 5 Preview: 41.2 (#112)
| Benchmark | Deepseek Coder v2 | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1181 | 1470 |
| ARC (AI2) Challenge | 64.3% | — |
Multimodal Not comparable
Deepseek Coder v2: —, Step 5 Preview: 41.0 (#41)
| Benchmark | Deepseek Coder v2 | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |
Multilingual Step 5 Preview leads
Deepseek Coder v2: 36.3 (#240), Step 5 Preview: 53.8 (#54)
| Benchmark | Deepseek Coder v2 | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1182 | 1432 |
| LMArena Chinese | 1201 | 1519 |
| LMArena Russian | 1188 | 1436 |
| LMArena Spanish | 1153 | 1439 |
| LMArena French | 1185 | — |
| LMArena German | 1164 | — |
| LMArena Japanese | 1126 | — |
| LMArena Korean | 1104 | — |
Instruction Following Step 5 Preview leads
Deepseek Coder v2: 61.7 (#242), Step 5 Preview: 76.0 (#50)
| Benchmark | Deepseek Coder v2 | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1180 | 1444 |
Long Context Step 5 Preview leads
Deepseek Coder v2: 37.0 (#224), Step 5 Preview: 44.5 (#64)
| Benchmark | Deepseek Coder v2 | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1219 | 1455 |
Writing & Preference Step 5 Preview leads
Deepseek Coder v2: 38.2 (#253), Step 5 Preview: 62.8 (#71)
| Benchmark | Deepseek Coder v2 | Step 5 Preview |
|---|---|---|
| LMArena Text | 1191 | 1442 |
| LMArena Creative Writing | 1120 | 1410 |
| LMArena Multi-Turn | 1177 | 1447 |
Frequently asked questions
Is Deepseek Coder v2 better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 35.9 on the Noometry Index.
Is Deepseek Coder v2 or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 38.1 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and Step 5 Preview share?
13 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Step 5 Preview has 18.