Step 5 Preview Adds Day-0 Support Across vLLM, SGLang, Transformers and Major Inference Providers

Key Info

StepFun's Step 5 Preview offers multiple deployment paths, including an OpenAI-compatible API served with vLLM or SGLang, offline inference with Transformers, and adaptation to multiple chip platforms and inference frameworks.

Highlights

• Serve an OpenAI-compatible API with vLLM or SGLang
• Run offline inference with Transformers
• Adapted for multiple chip platforms and inference frameworks
• Day-0 support from NVIDIA: try it on the NVIDIA platform, fine-tune with NeMo AutoModel, deploy with model-free NIM, and run on Blackwell
• Also available on Day 0 with Fireworks AI, Together Compute, Modal, DeepInfra and Featherless AI
• A single command is enough to start building