Video models · StepFun

Step-Video T2V (30B)

Name: Step-Video T2V (30B)
Author: StepFun

Backbone: 30.0B
Native output: 992×544, 204 frames
Default steps: 50
License: mit
Released: 2025-02

How to run it

Original weights · Hugging Face

Full-precision files from StepFun, for fine-tuning, vLLM, Transformers, Diffusers or ComfyUI.

huggingface.co/stepfun-ai/stepvideo-t2v (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from StepFun. The weights themselves download from Hugging Face.

github.com/stepfun-ai/Step-Video-T2V (opens in a new tab)

git clone https://github.com/stepfun-ai/Step-Video-T2V

Links checked 2 days ago against Hugging Face, Ollama and GitHub.

Memory needed

Peak GPU memory at the native output size. Offloading moves the text encoder, or the whole pipeline, to system RAM between steps.

Precision	Everything on GPU	Text encoder offloaded	Sequential offload	Download
`FP16`	78.2 GB	75.5 GB	35.5 GB	Original weights (opens in a new tab)
`FP8`	46.8 GB	45.5 GB	25.0 GB	Original weights (opens in a new tab)

Components

Backbone: 30B
VAE: 0.5B
Step-LLM: 1B
Hunyuan CLIP: 0.35B

Best hardware for Step-Video T2V (30B)

The least expensive GPU in our catalog that runs it without offloading is the Intel Arc Pro B70 (launch price $949) at FP8.

Quality	Hardware	Quant	Memory	Speed	Runs
76Great quality (A) · 76 of 100: how capable the model is at FP16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA RTX PRO 6000 Blackwell	`FP16`	78.2 GB / 96.0 GB	37.9 min	Slow
76Great quality (A) · 76 of 100: how capable the model is at FP16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GH200 Grace Hopper (96 GB GPU)	`FP16`	78.2 GB / 96.0 GB	36.0 min	Slow
76Great quality (A) · 76 of 100: how capable the model is at FP16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H100 NVL	`FP16`	78.2 GB / 94.0 GB	39.0 min	Slow
76Great quality (A) · 76 of 100: how capable the model is at FP16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA RTX PRO 6000 Blackwell Server Edition	`FP16`	78.2 GB / 96.0 GB	39.0 min	Slow
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H100 80GB SXM	`FP8`	46.8 GB / 80.0 GB	32.4 min	Slow
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H800 80GB SXM	`FP8`	46.8 GB / 80.0 GB	32.4 min	Slow
76Great quality (A) · 76 of 100: how capable the model is at FP16, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H200 141GB SXM	`FP16`	78.2 GB / 141 GB	36.0 min	Slow
76Great quality (A) · 76 of 100: how capable the model is at FP16, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H200 NVL	`FP16`	78.2 GB / 141 GB	39.0 min	Slow
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H100 80GB PCIe	`FP8`	46.8 GB / 80.0 GB	39.6 min	Slow
74Great quality (A) · 74 of 100: how capable the model is at FP8, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition	`FP8`	46.8 GB / 96.0 GB	37.5 min	Slow

Check it on your own hardware

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.