Model comparison
DeepSeek-V3.1 vs Step 3.5 Flash
DeepSeek-V3.1 and Step 3.5 Flash score almost the same on the Noometry Index (42.8 vs 42.3), so choose on price, context window or the category you care about most.
Last verified . 17 shared benchmarks.
DeepSeek
StepFunSummary
- They share 17 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 5 categories and Step 3.5 Flash in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where Step 3.5 Flash leads 42.8 to 36.3.
- Step 3.5 Flash is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.25 / $0.95 for DeepSeek-V3.1.
- Step 3.5 Flash accepts more context: 256K tokens versus 164K.
Side by side
| DeepSeek-V3.1 | Step 3.5 Flash | |
|---|---|---|
| Provider | DeepSeek | StepFun |
| Noometry Index | 42.8 | 42.3 |
| Released | 2025-08-21 | 2026-01-29 |
| Weights | Open | Open |
| Context window | 164K | 256K |
| Max output | 8K | 256K |
| Input $ / M tokens | $0.25 | $0.10 |
| Output $ / M tokens | $0.95 | $0.30 |
| Results tracked | 27 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3.5 Flash leads
DeepSeek-V3.1: 40.3 (#144), Step 3.5 Flash: 42.4 (#105)
| Benchmark | DeepSeek-V3.1 | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1417 | 1436 |
| WeirdML | 38.4% | — |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), Step 3.5 Flash: 22.2 (#202)
| Benchmark | DeepSeek-V3.1 | Step 3.5 Flash |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1411 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 28.4% |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
| Epoch Capabilities Index | 139.92 | — |
| ForecastBench | 58 | — |
Math Step 3.5 Flash leads
DeepSeek-V3.1: 38.9 (#122), Step 3.5 Flash: 42.6 (#84)
| Benchmark | DeepSeek-V3.1 | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1420 | 1408 |
| MathArena Final-Answer Competitions | — | 66.8% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Step 3.5 Flash: 39.6 (#132)
| Benchmark | DeepSeek-V3.1 | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1405 | 1421 |
| Vectara Hallucination Rate | 5.5% | — |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), Step 3.5 Flash: 50.5 (#119)
| Benchmark | DeepSeek-V3.1 | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1400 | 1385 |
| LMArena Chinese | 1469 | 1447 |
| LMArena French | 1447 | 1421 |
| LMArena German | 1411 | 1405 |
| LMArena Japanese | 1378 | 1354 |
| LMArena Korean | 1337 | 1352 |
| LMArena Russian | 1405 | 1385 |
| LMArena Spanish | 1431 | 1419 |
Instruction Following Too close to call
DeepSeek-V3.1: 73.9 (#110), Step 3.5 Flash: 73.1 (#124)
| Benchmark | DeepSeek-V3.1 | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1400 | 1385 |
Long Context Step 3.5 Flash leads
DeepSeek-V3.1: 36.3 (#232), Step 3.5 Flash: 42.8 (#117)
| Benchmark | DeepSeek-V3.1 | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1422 | 1402 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), Step 3.5 Flash: 58.8 (#113)
| Benchmark | DeepSeek-V3.1 | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1420 | 1403 |
| LMArena Creative Writing | 1401 | 1357 |
| LMArena Multi-Turn | 1408 | 1405 |
| EQ-Bench Creative Writing | 1436 | — |
Frequently asked questions
Is DeepSeek-V3.1 better than Step 3.5 Flash?
DeepSeek-V3.1 and Step 3.5 Flash score almost the same on the Noometry Index (42.8 vs 42.3), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-V3.1 or Step 3.5 Flash?
Step 3.5 Flash is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; DeepSeek-V3.1 lists at $0.25 and $0.95.
Is DeepSeek-V3.1 or Step 3.5 Flash better for coding?
Step 3.5 Flash scores higher on coding benchmarks: 42.4 versus 40.3 in the Noometry coding category.
Which has the bigger context window?
Step 3.5 Flash does, with 256K tokens against 164K.
How many benchmarks do DeepSeek-V3.1 and Step 3.5 Flash share?
17 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Step 3.5 Flash has 19.