Video models · HPC-AI Tech
Open-Sora 2.0 (11B)
- Backbone
- 11.0B
- Native output
- 768×768, 129 frames
- Default steps
- 50
- License
- apache-2.0
- Released
- 2025-03
How to run it
Original weights · Hugging Face
Full-precision files from HPC-AI Tech, for fine-tuning, vLLM, Transformers, Diffusers or ComfyUI.
Official repo · GitHub
Reference code, inference scripts and docs from HPC-AI Tech. The weights themselves download from Hugging Face.
github.com/hpcaitech/Open-Sora (opens in a new tab)
git clone https://github.com/hpcaitech/Open-SoraMemory 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 | 43.8 GB | 34.1 GB | 19.3 GB | Original weights (opens in a new tab) |
Components
- Backbone: 11B
- VAE: 0.25B
- T5-XXL: 4.76B
- CLIP-L: 0.123B
Best hardware for Open-Sora 2.0 (11B)
The least expensive GPU in our catalog that runs it without offloading is the AMD Radeon RX 7900 XT (launch price $899) at FP16.
| Quality | Hardware | Quant | Memory | Speed | Runs |
|---|---|---|---|---|---|
| 70Great quality (A) · 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA RTX PRO 6000 Blackwell |
FP16 | 43.8 GB / 96.0 GB | 36.6 min | Slow |
| 70Great quality (A) · 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA H100 80GB SXM |
FP16 | 43.8 GB / 80.0 GB | 34.8 min | Slow |
| 70Great quality (A) · 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA H800 80GB SXM |
FP16 | 43.8 GB / 80.0 GB | 34.8 min | Slow |
| 70Great quality (A) · 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA H100 NVL |
FP16 | 43.8 GB / 94.0 GB | 37.8 min | Slow |
| 70Great quality (A) · 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA GH200 Grace Hopper (96 GB GPU) |
FP16 | 43.8 GB / 96.0 GB | 34.8 min | Slow |
| 70Great quality (A) · 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA RTX PRO 6000 Blackwell Server Edition |
FP16 | 43.8 GB / 96.0 GB | 37.8 min | Slow |
| 70Great quality (A) · 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA H200 141GB SXM |
FP16 | 43.8 GB / 141 GB | 34.8 min | Slow |
| 70Great quality (A) · 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA H200 NVL |
FP16 | 43.8 GB / 141 GB | 37.8 min | Slow |
| 70Great quality (A) · 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. |  NVIDIA B200 |
FP16 | 43.8 GB / 180 GB | 30.5 min | Slow |
| 70Great quality (A) · 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. |  Apple M4 Max (64 GB) |
FP16 | 43.8 GB / 48.0 GB | 4.9 h | Very slowtight fit |
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.