Downloads · 30 days
8
12% of all-time downloads
hsr-video-lora-wan21
FT-LLM-2026-RAMEN/hsr-video-lora-wan21
hsr-video-lora-wan21 is a machine learning model from FT-LLM-2026-RAMEN. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft.
LoRA fine-tuned WAN2.1-T2V-1.3B for HSR robot manipulation video generation.
All-time downloads
68
Public
Repo size
378 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors189 MB · 100%
At a glance
- Library
- peft
- Access
- Public
- Created
- Mar 6, 2026
- Updated
- Mar 7, 2026
- SHA
- ca822ac4
Base models
Model card
From the Hugging Face model README
HSR Video LoRA - WAN2.1 T2V 1.3B
LoRA fine-tuned WAN2.1-T2V-1.3B for HSR robot manipulation video generation.
Training Details
- Base model: Wan-AI/Wan2.1-T2V-1.3B-Diffusers
- LoRA rank: 64, alpha: 64
- Training objective: Flow matching velocity prediction
- Training videos: 638 HSR robot episodes
- Epochs: 30, Final loss: 0.2015
- Target modules: to_q, to_k, to_v, to_out.0
Usage
from diffusers import WanPipeline
from peft import PeftModel
pipe = WanPipeline.from_pretrained("Wan-AI/Wan2.1-T2V-1.3B-Diffusers")
pipe.transformer = PeftModel.from_pretrained(pipe.transformer, "FT-LLM-2026-RAMEN/hsr-video-lora-wan21")
Project
ICRA 2026 RAMEN - DreamGen-inspired synthetic trajectory generation pipeline.
- Library
- peft
- Created
- Mar 6, 2026
- Updated
- Mar 7, 2026