Model comparison
DeepSeek-R1-Distill-Qwen-32B vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 35.5 on the Noometry Index.
Last verified . 0 shared benchmarks.
DeepSeek
StepFunSummary
- The widest gap is in reasoning, where Step 5 Preview leads 40.0 to 18.2.
- DeepSeek-R1-Distill-Qwen-32B has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-R1-Distill-Qwen-32B | Step 5 Preview | |
|---|---|---|
| Provider | DeepSeek | StepFun |
| Noometry Index | 35.5 | 47.9 |
| Released | 2025-01-20 | 2026-09-16 |
| Weights | Open | Proprietary |
| Context window | — | 1.02M |
| Max output | — | 66K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $2.70 |
| Results tracked | 14 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
DeepSeek-R1-Distill-Qwen-32B: 36.1 (#212), Step 5 Preview: 51.8 (#36)
| Benchmark | DeepSeek-R1-Distill-Qwen-32B | Step 5 Preview |
|---|---|---|
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
| BigCodeBench Instruct | 43.9% | — |
| LiveBench Coding | 33.7% | — |
| LMArena Coding | — | 1480 |
| BigCodeBench Complete | 54.9% | — |
Agentic & Tool Use Not comparable
DeepSeek-R1-Distill-Qwen-32B: 28.1 (#94), Step 5 Preview: —
| Benchmark | DeepSeek-R1-Distill-Qwen-32B | Step 5 Preview |
|---|---|---|
| BALROG | 19.5% | — |
Reasoning Step 5 Preview leads
DeepSeek-R1-Distill-Qwen-32B: 18.2 (#284), Step 5 Preview: 40.0 (#57)
| Benchmark | DeepSeek-R1-Distill-Qwen-32B | Step 5 Preview |
|---|---|---|
| CritPt | — | 20.9% |
| Chess Puzzles | 1% | — |
| LiveBench Reasoning | 52.3% | — |
| LMArena Hard Prompts | — | 1465 |
| LiveBench Data Analysis | 45.4% | — |
| Epoch Capabilities Index | 137.44 | — |
| LiveBench | 45.5% | — |
Math Step 5 Preview leads
DeepSeek-R1-Distill-Qwen-32B: 34.5 (#194), Step 5 Preview: 46.1 (#72)
| Benchmark | DeepSeek-R1-Distill-Qwen-32B | Step 5 Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 55.6% | — |
| ProofBench | — | 42% |
| LiveBench Math | 59.4% | — |
| LMArena Math | — | 1470 |
Knowledge Step 5 Preview leads
DeepSeek-R1-Distill-Qwen-32B: 35.7 (#182), Step 5 Preview: 41.2 (#112)
| Benchmark | DeepSeek-R1-Distill-Qwen-32B | Step 5 Preview |
|---|---|---|
| GPQA Diamond | 64.1% | — |
| LMArena Expert | — | 1470 |
Multimodal Not comparable
DeepSeek-R1-Distill-Qwen-32B: —, Step 5 Preview: 41.0 (#41)
| Benchmark | DeepSeek-R1-Distill-Qwen-32B | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |
Multilingual Not comparable
DeepSeek-R1-Distill-Qwen-32B: —, Step 5 Preview: 53.8 (#54)
| Benchmark | DeepSeek-R1-Distill-Qwen-32B | Step 5 Preview |
|---|---|---|
| LMArena Non-English | — | 1432 |
| LMArena Chinese | — | 1519 |
| LMArena Russian | — | 1436 |
| LMArena Spanish | — | 1439 |
Instruction Following Step 5 Preview leads
DeepSeek-R1-Distill-Qwen-32B: 61.6 (#243), Step 5 Preview: 76.0 (#50)
| Benchmark | DeepSeek-R1-Distill-Qwen-32B | Step 5 Preview |
|---|---|---|
| LiveBench Instruction Following | 55.7% | — |
| LMArena Instruction Following | — | 1444 |
Long Context Not comparable
DeepSeek-R1-Distill-Qwen-32B: —, Step 5 Preview: 44.5 (#64)
| Benchmark | DeepSeek-R1-Distill-Qwen-32B | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | — | 1455 |
Writing & Preference Step 5 Preview leads
DeepSeek-R1-Distill-Qwen-32B: 49.6 (#188), Step 5 Preview: 62.8 (#71)
| Benchmark | DeepSeek-R1-Distill-Qwen-32B | Step 5 Preview |
|---|---|---|
| LMArena Text | — | 1442 |
| LMArena Creative Writing | — | 1410 |
| LMArena Multi-Turn | — | 1447 |
| LiveBench Language | 26.8% | — |
Frequently asked questions
Is DeepSeek-R1-Distill-Qwen-32B better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 35.5 on the Noometry Index.
Is DeepSeek-R1-Distill-Qwen-32B or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 36.1 in the Noometry coding category.
How many benchmarks do DeepSeek-R1-Distill-Qwen-32B and Step 5 Preview share?
0 benchmarks have published results for both models. DeepSeek-R1-Distill-Qwen-32B has 14 scored results on Noometry and Step 5 Preview has 18.