Author: Juhi Singh | HuggingFace
This is Checkpoint 1600 in our LCM-LoRA training series. Each checkpoint has different characteristics:
Checkpoint 400 • Checkpoint 800 • Checkpoint 1200 • Checkpoint 1600 (current)
This checkpoint represents training at 1600 steps in our LCM-LoRA progression for Stable Diffusion v1.5.
Characteristics:
Key Features:
This checkpoint is part of a training series. Compare with other checkpoints:
| Steps | Model | Characteristics |
|---|---|---|
| 400 | lcm-lora-sd1.5-400 | Early training checkpoint showing foundational LCM capabilities. Provides decent... |
| 800 | lcm-lora-sd1.5-800 | Mid-training checkpoint with vibrant, artistic outputs. Strong visual impact wit... |
| 1200 | lcm-lora-sd1.5-1200 | Higher training with more refined outputs. Some prompts may show signs of overfi... |
| 1600 | lcm-lora-sd1.5-1600 | Final training checkpoint with mature, consistent outputs. Well-balanced and rel... → This checkpoint |
Compare training progression and characteristics:
| Steps | Model Link | Style | Speed (RTX 3090) | Best For |
|---|---|---|---|---|
| 400 | lcm-lora-sd1.5-400 | Soft, baseline | 2-3s @ 6 steps | Fast experimentation, understanding earl... |
| 800 | lcm-lora-sd1.5-800 | Vibrant, saturated | 2-3s @ 6 steps | Artistic applications, vibrant aesthetic... |
| 1200 | lcm-lora-sd1.5-1200 | Balanced, natural | 2-3s @ 6 steps | Balanced colors, natural tones, specific... |
| 1600 | lcm-lora-sd1.5-1600 | Mature, consistent | 2-3s @ 6 steps | Most training, consistent results, produ... ← Current |
See how outputs evolve across the training series. Each grid shows: Baseline SD1.5 (50 steps) vs LCM-LoRA at 2, 4, and 6 steps.
pip install --upgrade diffusers transformers accelerate
import torch
from diffusers import StableDiffusionPipeline, LCMScheduler
# Load base SD1.5 model
pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
torch_dtype=torch.float16
)
pipe.to("cuda")
# Load this LCM-LoRA checkpoint
pipe.load_lora_weights("Mercity/lcm-lora-sd1.5-1600")
# IMPORTANT: Use LCM scheduler
pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config)
# Generate with just 4-6 steps!
prompt = "a portrait of a cat wearing a detective hat, film noir style"
image = pipe(
prompt=prompt,
num_inference_steps=6,
guidance_scale=1.0
).images[0]
image.save("output.png")
num_inference_steps = 6 # Optimal for this checkpoint
guidance_scale = 1.0 # Required for LCM
| Parameter | Value |
|---|---|
| Checkpoint | 1600 |
| Base Model | runwayml/stable-diffusion-v1-5 |
| Training Steps | 1600 |
| Dataset | Mercity/laion-subset |
| LoRA Rank | 96 |
| LoRA Alpha | 96 |
| Resolution | 512×512 |
| Batch Size | 64 |
| Learning Rate | 1e-4 |
| Optimizer | AdamW |
The comparison grid above shows outputs from this checkpoint at 2, 4, and 6 inference steps, compared to standard SD1.5 at 50 steps.
Prompts included:
All sample images for this checkpoint are available in the samples/ directory.
To test generalization beyond the training distribution, we generated images for 5 OOD prompts that are deliberately different from training prompts:
🐠 Underwater Scene
🚀 Space/Astronomy
🍰 Food Photography
👴 Human Portrait
🎨 Abstract Art
Why OOD Validation? These prompts test whether the model truly learned general concepts rather than just memorizing training prompts. Good OOD performance indicates robust generalization.
All validation images can be found in the validation/ directory. See validation/prompts.txt for the complete list of prompts used.
| Method | Steps | Time (A100) | Time (RTX 3090) |
|---|---|---|---|
| SD1.5 Default | 50 | ~15s | ~25s |
| SD1.5 Fast | 25 | ~8s | ~13s |
| LCM-LoRA (this) | 6 | ~2s | ~3s |
| LCM-LoRA (this) | 4 | ~1.5s | ~2s |
This checkpoint is part of a training series showing LCM-LoRA evolution:
Training Steps: 400 ──→ 800 ──→ 1200 ──→ 1600
↓ ↓ ↓ ↓
Quality: Baseline Peak Refined Mature
Style: Soft Vibrant Balanced Stable
# Download all checkpoints for comparison
huggingface-cli download Mercity/lcm-lora-sd1.5-400
huggingface-cli download Mercity/lcm-lora-sd1.5-800
huggingface-cli download Mercity/lcm-lora-sd1.5-1200
huggingface-cli download Mercity/lcm-lora-sd1.5-1600
LCMScheduler - Required for LCMguidance_scale=1.0 - CFG doesn't work with LCMLCMScheduler - other schedulers won't workguidance_scale must be 1.0 (CFG incompatible with LCM)If you use this model in your research, please cite:
@article{luo2023latent,
title={Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference},
author={Luo, Simian and Tan, Yiqin and Huang, Longbo and Li, Jian and Zhao, Hang},
journal={arXiv preprint arXiv:2310.04378},
year={2023}
}
@article{hu2021lora,
title={LoRA: Low-Rank Adaptation of Large Language Models},
author={Hu, Edward J and Shen, Yelong and Wallis, Phillip and Allen-Zhu, Zeyuan and Li, Yuanzhi and Wang, Shean and Wang, Lu and Chen, Weizhu},
journal={arXiv preprint arXiv:2106.09685},
year={2021}
}
This model is released under the same license as Stable Diffusion v1.5:
Base model
runwayml/stable-diffusion-v1-5