Loading benchmark catalogue
Compare performance benchmarks, coding intelligence, reasoning evaluations, context limits, and API economics between Step 5 Preview (StepFun) and DeepSeek V4.1 Flash 1017 (DeepSeek).
| Metric / Capability | Step 5 Preview | DeepSeek V4.1 Flash 1017 |
|---|---|---|
| Developer / Organization | StepFun | DeepSeek |
| Release Date | 2026-10-08 | 2026-10-17 |
| Deployment / Weights | Open Weights | Open Weights |
| Context Window | 1.0M tokens (1,048,576 tokens) | 1.0M tokens (1,048,576 tokens) |
| Input Pricing (/1M Tokens) | Free (Open Weights) | Free (Open Weights) |
| Output Pricing (/1M Tokens) | Free (Open Weights) | Free (Open Weights) |
| Benchmark / Domain | Step 5 Preview | DeepSeek V4.1 Flash 1017 | Advantage |
|---|---|---|---|
Artificial Analysis Intelligence Index Composite index aggregating 10+ evaluations of general reasoning and knowledge. | 43.7% | 39.5% | Step 5 Preview (+11%) |
Artificial Analysis Coding Agent Index v2.0 Composite index of coding-agent systems (model + harness) across FrontierCode v1.1, SWE-Atlas-QnA, and Terminal-Bench v4. | 43.1% | 46.3% | DeepSeek V4.1 Flash 1017 (+7%) |
Terminal-Bench 4.0 Sandboxed command-line tasks requiring shell and file-system operation. | 33.3% | 42.3% | DeepSeek V4.1 Flash 1017 (+27%) |
GPQA Diamond Graduate-level physics, chemistry, and biology questions written by domain experts. | 93.5% | 90.9% | Step 5 Preview (+3%) |
Humanity's Last Exam Expert-written questions across hundreds of disciplines designed to resist saturation. | 46.5% | 42.6% | Step 5 Preview (+9%) |
FrontierMath v2 (Tiers 1-3) Unpublished, research-level mathematics problems across Tiers 1-3. | — | 67.4% | — |
Design Arena ELO Human-preference ELO across UI, 3D, data-visualization, and game-dev generations. | — | 1362 | — |
Step 5 Preview leads across 3 evaluated benchmarks compared to 2 for DeepSeek V4.1 Flash 1017, showing notable advantages in Artificial Analysis Intelligence Index, GPQA Diamond, Humanity's Last Exam.
Both models feature equivalent context windows of 1.0M tokens (1,048,576 tokens).
Both models have comparable pricing structures or open weight deployment options.
Step 5 Preview is an open-weights model. DeepSeek V4.1 Flash 1017 is an open-weights model.
Analyze 130+ models side-by-side using interactive radar charts, cost-efficiency quadrant plots, and live dataset filters on the full NeuBench index.
Open Interactive Comparator