StepFun: Step 5 Preview
by StepFunStepFun: Step 5 Preview is a large language model from StepFun. It costs $1.00 per million input tokens and $2.70 per million output tokens. Its context window is 1.0M tokens.
- Input / 1M tokens
- $1.00
- Output / 1M tokens
- $2.70
- Cached input / 1M
- $0.100
- Context window
- 1.0M
On repeated prefixes
tokens
Who serves it cheapest
4 hosts serve Step 5 Preview. Same weights, same API - the price difference is pure margin and routing.
| Provider | Input / 1M | Output / 1M | Context | Throughput | Uptime 24h |
|---|---|---|---|---|---|
| InferenceNetCheapestfp4 | $1.00 | $2.70 | 1.0M | - | 100.0% |
| Featherlessfp4 | $1.00 | $2.70 | 262K | - | 57.9% |
| DeepInfrafp8 | $1.00 | $2.70 | 1.0M | - | 100.0% |
| StepFunfp8 | $1.00 | $2.70 | 1M | - | 100.0% |
The spread between InferenceNet and StepFun is about the same for identical weights. Quantization and context limits differ, so check both columns before switching.
Benchmarks
Independent scores published alongside the catalogue.
- Intelligence index
- 43.7
About Step 5 Preview
Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters). It performs strongly in software engineering and professional...
Specifications
| Model ID | stepfun/step-5-preview |
|---|---|
| Provider | StepFun |
| Context window | 1.0M tokens |
| Max output | 64K tokens |
| Input modalities | text, image, video |
| Output modalities | text |
| Knowledge cutoff | - |
| Open weights | Yes - stepfun-ai/Step-5-Preview |
| Released | October 8, 2026 |
Cheaper alternatives
Models that cost less than Step 5 Preview while keeping at least half its context window and every input modality it supports.
Frequently asked
How much does StepFun: Step 5 Preview cost?
$1.00 per million input tokens and $2.70 per million output tokens. Cached input reads cost $0.100 per million tokens.
What is the context window of StepFun: Step 5 Preview?
1.0M tokens, with up to 64K tokens of output per request.
Which provider serves StepFun: Step 5 Preview cheapest?
InferenceNet at $1.00 per million input tokens - about the same cheaper than StepFun, the most expensive of the 4 hosts serving it.