stepfun
Step 5 Preview
Step 5 Preview is StepFun’s flagship AI model for agentic coding, professional knowledge work, and financial analysis. Built on a sparse Mixture-of-Experts architecture with 600B total parameters and 27B active per token, it combines a 1M-token context window with vision input. From building interactive applications and debugging code to conducting research and producing analytical reports, it supports complex workflows that require sustained reasoning, tool use, and iterative refinement.
implicit cachingreasoningstructured outputtool usevideo inputvisiontextimagevideo
Overview
- Type
- language
- Context
- 1M
- Max output
- 1M
- Input / output
- text, image, video / text
API
Model identifier
stepfun/step-5-previewProviders and live performance
Rolling Gateway metrics. Latency is time to first token; throughput is output tokens per second.
| Provider | Status | Latency | Throughput | Uptime 15m / 1h / 1d | Input | Output | Cache read / write | Discount | Regions | ZDR | No training |
|---|---|---|---|---|---|---|---|---|---|---|---|
| deepinfra | Operational | 0.48s / 5.39s | 129.0 / 141.8 tok/s | 100.00% / 96.00% / 96.56% | $0.95/M | $2.56/M | $0.05/M / — | — | — | Yes | Yes |
| stepfun | Operational | 1.89s / 3.49s | 81.0 / 81.9 tok/s | 100.00% / 100.00% / 99.97% | $1.00/M | $2.70/M | $0.05/M / — | — | — | No | No |
Pricing
input
0.00000095
output
0.000002565
input cache read
0.0000000475
varies by provider
Yes
Capabilities
implicit cachingreasoningstructured outputtool usevideo inputvisiontextimagevideo
Supported parameters
max tokens
Yes
temperature
Yes
stop
Yes
tools
Yes
tool choice
Yes
reasoning
Yes
include reasoning
Yes
response format
Yes
structured outputs
Yes
Reasoning options
options
{
"type": "effort",
"values": [
"low",
"medium",
"high"
]
}
Metadata
type
language
context window
1M
max output
1M
released
Sep 20, 2026
temperature control
Yes
zero data retention
some
no training
some
api specifications
v2, v3, v4