Will It Run AI

Helios 14B

Frontier

by BestWishYsh

14B distilled video model achieving 19.5 FPS real-time generation. Based on Wan2.1-T2V-14B with pyramid distillation for minute-scale coherent video. Apache 2.0 licensed.

  • 19.5 FPS real-time video generation
  • Minute-scale coherent video
  • Distilled from Wan2.1-T2V-14B
  • Apache 2.0 — fully open for commercial use

Your hardware

Detecting...

Parameters14B
Max Resolution1280×720
Max Frames240
FPS24
Architecture3D-DIT
Licenseapache-2.0

Image Quality Benchmarks

Measured quality metrics for Helios 14B outputs.

Human Preference Score85%

How often humans prefer this model's output (0-100%)

Aesthetic Score7.8

Visual quality and composition rating (5-9 scale)

This model requires 46+ GB VRAM for basic video generation. A GPU with 24GB+ VRAM is recommended.

VRAM by Scenario

VRAM estimates at FP16 and FP8 precision. FP8 uses ~40% less memory with minimal quality loss. Grade shows how well each GPU handles the generation workload.

FP16 (full precision)

ScenarioVRAMRTX 4090 24GBRTX 3060 12GBRTX 4060 8GBMacBook Pro M4 Pro 24GB
512×512 · 25 frames44.2 GBFFFF
768×512 · 25 frames46.3 GBFFFF
768×512 · 100 frames52.6 GBFFFF
1280×720 · 25 frames54.8 GBFFFF

FP8 (quantized — ~40% less VRAM)

ScenarioVRAMRTX 4090 24GBRTX 3060 12GBRTX 4060 8GBMacBook Pro M4 Pro 24GB
512×512 · 25 frames24.6 GBBFFF
768×512 · 25 frames26.7 GBBFFF
768×512 · 100 frames33.0 GBDFFF
1280×720 · 25 frames35.1 GBFFFF

Optimization Tips

Turbo / LCM distillation

Use distilled scheduler at 4-8 steps for faster iteration

Run with Python

Run with Python (diffusers)
from diffusers import WanPipeline
import torch

pipe = WanPipeline.from_pretrained(
    "BestWishYsh/Helios-Distilled",
    torch_dtype=torch.float16
)
pipe.to("cuda")

frames = pipe(
    prompt="your prompt here",
    num_inference_steps=20,
    guidance_scale=5.0,
    num_frames=240,
).frames[0]
# Save frames or export as video

Get started

Setup instructions for running Helios 14B locally

1. Download the model

Get the checkpoint from HuggingFace

2. Place in:

ComfyUI/models/checkpoints/

3. Launch ComfyUI

python main.py
Note: Video generation requires video output nodes. Install ComfyUI-VideoHelperSuite from the ComfyUI Manager for SaveAnimatedWEBP or VHS_VideoCombine nodes.

Memory Breakdown

VRAM allocation for 25 frames at 768×512 on RTX 4090 24GB

Required: 46.3 GBAvailable: 24.0 GB
Weights28.0 GB
VAE0.2 GB
Text Encoder9.4 GB
Activations6.0 GB
Overhead0.5 GB

Estimated Generation Time

25 frames at 768×512, 30 steps, FP16.

RTX 4090 24GB~3m 10s
RTX 3060 12GB~11m 55s
RTX 4060 8GB~17m 55s
MacBook Pro M4 Pro 24GB~25m 30s

Sample Outputs

Available Formats & Downloads

Download Helios 14B in different precisions. Lower precision = less VRAM but slight quality loss.

格式精度大小提供商
safetensors推荐BF1628.0 GBofficial下载

LoRA Ecosystem

Limited

Compatible with some Wan 2.1 LoRAs.

Related Workflows

You might also like

Frequently asked questions

FAQ — Helios 14B

How much VRAM does Helios 14B need for video?

Helios 14B (14B parameters) requires approximately 46.3 GB of VRAM at FP16 precision for generating 25 frames at 768×512. Video generation typically requires more VRAM than image generation due to temporal attention layers.

Can I run Helios 14B on RTX 4090?

Helios 14B exceeds the RTX 4090's 24 GB VRAM at FP16 for video generation. Consider reducing resolution, frame count, or using a GPU with more VRAM.

How long does it take to generate a video with Helios 14B?

On a reference GPU (RTX 4090 24GB), Helios 14B generates a 25-frame video at 768×512 in approximately ~3m 10s at FP16 with 30 inference steps. Faster GPUs with higher memory bandwidth will reduce generation time.

What resolution and frame count does Helios 14B support?

Helios 14B supports up to 1280×720 resolution and 240 frames per generation at 24 FPS. Higher resolutions and frame counts require proportionally more VRAM.

About Helios 14B

Use cases
video-generationreal-time-videolong-video
Recommended runtimes
comfyuidiffusers

See also