Clover-Image-Tiny
Collection
A tiny image-generation model ecosystem with LoRA support and private, offline Core ML deployment on Apple devices. • 11 items • Updated
Clover Image Tiny — Pointillism LoRA
A rank-16 style LoRA trained on
neonforestmist/GPT_Pointillism_Style_Images
for neonforestmist/Clover-Image-Tiny.
The corresponding stateful Core ML adapter is neonforestmist/clover-image-tiny-pointillism-lora-coreml, which targets neonforestmist/Clover-Image-Tiny-CoreML.
Examples
The prompt set used for this gallery is preserved at
examples/prompt-gallery/prompts.txt.
- Prompt
- pointillism painting, leaves of different color
- Prompt
- pointillism painting, a mountains scene
- Prompt
- pointillism painting, a blue and green bird
- Prompt
- pointillism painting, a pink lotus in a swamp
- Prompt
- pointillism painting, a butterfly
- Prompt
- pointillism painting, ocean tides at night
- Prompt
- pointillism painting, a toucan
- Prompt
- pointillism painting, a blossom forest
- Prompt
- pointillism painting, a person looking at the night sky
Base model → Pointillism LoRA
Same prompt, same greenhouse subject, with the style adapter applied:
The original four validation samples remain available in the repository as
image_0.png through image_3.png.
Usage
Use the prompt trigger pointillism painting.
import torch
from diffusers import DiffusionPipeline
pipe = DiffusionPipeline.from_pretrained(
"neonforestmist/Clover-Image-Tiny",
torch_dtype=torch.float16,
).to("cuda")
pipe.load_lora_weights(
"neonforestmist/clover-image-tiny-pointillism-lora"
)
image = pipe(
"pointillism painting, a small blue cat in a flower garden",
num_inference_steps=20,
guidance_scale=7.5,
).images[0]
Training
- Base revision:
63b0e9f6be9c00888ff464f342a9ef052bf76681 - Dataset revision:
0f1bbd4c768ed9bd65a62ae7f7bf3f0aee351691 - Resolution: 512 × 512
- Optimizer steps: 700
- Rank: 16
- Batch size: 1
- Learning rate: 1e-4 with cosine decay and 70 warmup steps
- Min-SNR gamma: 5
- Precision: fp16
- Seed: 20260730
- Trainer: Diffusers 0.39.0
train_text_to_image_lora.py
The reproducible job configuration is included in the Clover source
repository under training/.
Provenance
- Base model:
neonforestmist/Clover-Image-Tiny - Training dataset:
neonforestmist/GPT_Pointillism_Style_Images - Core ML base:
neonforestmist/Clover-Image-Tiny-CoreML - Training and conversion source:
neonforestmist/clover-image-tiny-lora-trainer
License and limitations
These adapter weights are a derivative of Clover Image Tiny and use the CreativeML Open RAIL-M license. The training dataset is Apache-2.0. Generated content can inherit limitations and biases from the base checkpoint and training data; review outputs before use.
- Downloads last month
- 27