stepfun/step-5-preview
stepfun/step-5-preview is a advanced-tier language model from RouterPlus. stepfun/step-5-preview is a proprietary model available exclusively through RouterPlus's API. The Step family is RouterPlus's model series for this capability tier.
stepfun/step-5-preview is priced at $1/1M input tokens and $2.7/1M output tokens. Output tokens are 2.7× more expensive than input tokens — this ratio reflects the sequential nature of autoregressive generation, which requires a separate forward pass per output token versus parallel processing of all input tokens simultaneously. At a typical usage pattern of 1,000 input tokens and 500 output tokens per call, stepfun/step-5-preview costs approximately $2.35 per 1,000 API calls. Prompt caching reduces repeated input tokens to $0.05/1M — a 95% discount that compounds significantly for applications with long shared system prompts, multi-turn conversations, or document Q&A where the same context is reused across many queries.
Its 0-token context window is standard for this tier — adequate for most chat and completion tasks, but may require chunking for long-document workloads.
stepfun/step-5-preview is best suited to general language tasks. Supported capabilities include streaming token output for real-time UIs. When evaluating this model for a specific workload, the most important factors are: context window (does your input fit?), output price (how long are your generations?), and latency (is this interactive or batch?). The pricing table above shows how stepfun/step-5-preview compares to similar models across these dimensions.
stepfun/step-5-preview sits in the advanced tier — strong general-purpose models balancing capability and cost, suitable for most production applications. Unlike open-weights alternatives, stepfun/step-5-preview is only available via RouterPlus's API — there is no self-hosted option, which simplifies deployment but creates vendor dependency. When comparing stepfun/step-5-preview to alternatives, use the similar models section on this page to find models at comparable price points with different capability tradeoffs. The intelligence score (N/A/100) reflects aggregate benchmark performance — a useful proxy for general capability, though real-world performance varies significantly by task type. Always test with your specific prompts and use case before committing to a model for production.
Input price vs 184 advanced models tracked. Avg: $1.38/1M.
| Model ID | stepfun/step-5-preview |
| Provider | RouterPlus |
| Model Family | Step |
| Tier | Advanced |
| Context Window | 0 tokens (0) |
| Max Output Tokens | — |
| Input Price | $1 / 1M tokens |
| Output Price | $2.7 / 1M tokens |
| Cached Input Price | $0.05 / 1M tokens |
| Throughput | — |
| Latency (P50) | — |
| Intelligence Score | — |
| LiveBench Score | — |
| Cost per Task (LiveBench) | — |
| Data Confidence | — |
| HLE Score | — |
| GPQA Diamond | — |
| SWE-bench Score | — |
| OSWorld Score | — |
| BrowseComp Score | — |
| Terminal-Bench Score | — |
| AutoBench Score | — |
| Parameters | — |
| Vision Support | No |
| Function Calling | No |
| Streaming | Yes |
| Reasoning Model | No |
| Open Weights | No |
| Data Source | Static (verified) |
| Last Updated | 2026-10-28 |
Self-hosting stepfun/step-5-preview? These GPUs are well-matched for this model's compute and memory requirements.