Pricing
| Provider | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|
| RunInfra | $0.90 | $0.09 | $2.43 |
USD, as of November 12, 2026. Live prices

| Provider | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|
| RunInfra | $0.90 | $0.09 | $2.43 |
USD, as of November 12, 2026. Live prices
| Workload | RunInfra |
|---|---|
| 1,000 chat replies2K in, 500 out each | $3.02 |
| 1,000 agent tasks50K in, 5K out each | $57.15 |
| Monthly batch job500M in, 50M out | $571.50 |
curl https://api.routerplus.com/v1/chat/completions \
-H "Authorization: Bearer $ROUTERPLUS_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "stepfun/step-5-preview",
"messages": [{ "role": "user", "content": "Hello" }]
}'On RouterPlus, Step 5 Preview costs $0.90 per million input tokens and $2.43 per million output tokens at RunInfra, the lowest listed prices and the provider where routing starts. Prices as of November 12, 2026.
RunInfra serves Step 5 Preview on RouterPlus today.
Use "stepfun/step-5-preview" as the model with the OpenAI-compatible endpoint at https://api.routerplus.com/v1. The same key also accepts the Anthropic API format.
1.05M tokens of context and up to 32.8K output tokens, per the RouterPlus catalog. A provider can list a lower limit; the console shows each one.
None of the 1 providers serving Step 5 Preview on RouterPlus list prompt retention.
Yes. Tell us the model, the throughput you need and the region with "I need inference", and RouterPlus lines up reserved capacity for it.