#6060th of 75 in Image and videoFor image, video and design, video and animation
hunyuanvideo-1.5
8.3B text-to-video and image-to-video diffusion model with training code
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.6k
- License
- custom license
- Last commit
- Apr 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
HunyuanVideo-1.5 is Tencent's video generation model: an 8.3B-parameter DiT with a 3D causal VAE, producing text-to-video and image-to-video at 480p and 720p, with super-resolution models to 1080p. The repo includes generate.py inference (multi-GPU via torchrun), step- and CFG-distilled checkpoints, caching options, and a train.py with FSDP and LoRA support using the Muon optimizer.
Who it is for: Developers and researchers generating or fine-tuning video models on NVIDIA GPUs
Strengths
- Runs with as little as 14 GB GPU memory when model offloading is enabled
- Distilled variants: 480p I2V step-distilled in 8 or 12 steps, plus CFG-distilled checkpoints
- Includes training code (FSDP, context parallel) and LoRA tuning, not just inference
- Integrates with ComfyUI, Diffusers and LightX2V
Weaknesses
- Requires an NVIDIA CUDA GPU on Linux; no CPU or Mac path described
- Some checkpoints still marked coming soon, e.g. 720p T2V CFG-distill and sparse T2V
- Prompt rewriting expects a separate vLLM endpoint serving a large Qwen3 model
- Optional speedups need extra builds: Flash Attention, SageAttention, flex-block-attn, sgl-kernel
What it needs
- GPU required
- Needs PyTorch, CUDA, Flash Attention, SageAttention, flex-block-attn, sgl-kernel, vLLM (optional, prompt rewrite)
- Models: HunyuanVideo-1.5 480p/720p T2V, HunyuanVideo-1.5 480p/720p I2V, 720p and 1080p super-resolution, Diffusers
Also in Image and video
See all 75| Rank | Project | Score |
|---|---|---|
| 1 | frigateLocal NVR with realtime object detection for IP cameras | 93 out of 100 |
| 2 | MoneyPrinterTurboGenerates short videos from a topic, with script, footage, voiceover and subtitles | 88 out of 100 |
| 3 | toonflow-appInfinite-canvas tool for AI short drama and comic video production | 87 out of 100 |
| #4 | InvokeAIWeb UI for diffusion image generation with inpainting canvas and node workflows | 86 out of 100 |
| #5 | ComfyUINode-graph editor for running diffusion image, video, audio and 3D workflows | 85 out of 100 |