CivArchive
    Preview 1
    Preview 2
    Preview 3
    Preview 4
    Preview 5
    Preview 6
    Preview 7
    Preview 8
    Preview 9
    Preview 10
    Preview 11
    Preview 12
    Preview 13
    Preview 14
    Preview 15
    Preview 16
    Preview 17
    Preview 18
    Preview 19

    Weekly MidJourney Style Experiment: Last 500 Images

    Update (11/02/23) : adding WEEK 3 FFUsion (ALL IN) LoRA of Hot, Rising, and New

    ๐Ÿ—“ Current Week's Exploration (10.31.23): This week LoRAs are fine-tuned for the

    ๐Ÿ†'Style Capture & Fusion Showdown' (rejected entry)

    Week 3 Styles of Hot, Rising, and New MJ categories: 10.31.23

    ๐Ÿ”ฅ Previous Weekโ€™s LoRAs (10.23.23):
    last week styles from the 'Hot', 'Rising', and 'New' category



    A recent experiment conducted using the last 500 images from the 'MidJourney' categories: Hot, Rising, and New.

    Data Acquisition and Integrity:

    • All images were sourced responsibly, with no use of unofficial tools for acquisition.

    • The images were obtained using the sanctioned Corporate/Enterprise account.

    Technical Overview:

    • The images were processed using the ViT-L-14/openai model (quick sloppy run). For testing, the prodigy tool was employed.

    • It's important to note that the current quality of results does not align with our typical production standards. However, for those interested in further details, a training set from the official Civitai trainer is available.

    • The experiment utilized the capabilities of the Civitai trainer(default out of the box configuration)


    09/26/2023 03:48:12 AM

    SUBMITTED

    09/26/2023 03:48:40 AM

    PROCESSING

    09/26/2023 05:21:46 AM

    READY

    • Lora FA text encoder, and the Kohya tools, all operating on the H100 80GB.

    We appreciate your continued interest and support. Further updates will be provided as the experiment progresses.

    Each one took 20-30min


    ๐Ÿ“‚ MidJ_Last_500_-_Experiment.safetensors

    • ๐Ÿ“… Date: 2023-09-26T02:20:38

    • ๐Ÿท๏ธ Title: MidJ_Last_500_-_Experiment

    • ๐Ÿ–ผ๏ธ Resolution: 1024x1024

    • ๐Ÿงช Architecture: stable-diffusion-xl-v1-base/lora

    • ๐ŸŒ Network Dimensions:

      • Dim/Rank: 32.0

      • Alpha: 16.0

    • ๐Ÿ”Œ Module: networks.lora

    • ๐Ÿ”ง Configurations:

      • Learning Rate: 0.0005

      • UNet LR: 0.0005

      • TE LR: 5e-05

      • Optimizer: bitsandbytes.optim.adamw.AdamW8bit(weight_decay=0.1)

      • Scheduler: cosine_with_restarts

      • Warmup Steps: 0

      • Epochs: 10

      • Batches per Epoch: 128

      • Gradient Accumulation Steps: 1

      • Train Images: 500

      • Regularization Images: 0

      • Multires Noise Iterations: 6.0

      • Multires Noise Discount: 0.3

      • Min SNR Gamma: 5.0

      • Zero Terminal SNR: True

      • Max Gradient Norm: 1.0

      • Clip Skip: 1

      • Dataset Directories: 1

      • Image Count: 500 images

    • ๐Ÿ“ˆ Stats:

      • UNet Weight (Avg. Magnitude): 3.0170

      • UNet Weight (Avg. Strength): 0.0111

      • Text Encoder (1) - Weight (Avg. Magnitude): 1.7304

      • Text Encoder (1) - Weight (Avg. Strength): 0.0087

      • Text Encoder (2) - Weight (Avg. Magnitude): 1.7614

      • Text Encoder (2) - Weight (Avg. Strength): 0.0068

    ๐Ÿ“‚ FF-Midj-Last-v0563.safetensors

    • ๐Ÿ“… Date: 2023-09-26T01:16:09

    • ๐Ÿท๏ธ Title: FF-Midj-Last-v0563

    • ๐Ÿ–ผ๏ธ Resolution: 1024x1024

    • ๐Ÿงช Architecture: stable-diffusion-xl-v1-base/lora

    • ๐ŸŒ Network Dimensions:

      • Dim/Rank: 64.0

      • Alpha: 32.0

    • ๐Ÿ”Œ Module: networks.lora

    • ๐Ÿ“ˆ Stats:

      • UNet Weight (Avg. Magnitude): 2.6731

      • UNet Weight (Avg. Strength): 0.0076

      • Text Encoder (1) - Weight (Avg. Magnitude): 2.5809

      • Text Encoder (1) - Weight (Avg. Strength): 0.0091

      • Text Encoder (2) - Weight (Avg. Magnitude): 2.6613

      • Text Encoder (2) - Weight (Avg. Strength): 0.0072


    ๐Ÿ“‚ FF-Midj-Rise-v0564.safetensors

    • ๐Ÿ“… Date: 2023-09-26T02:01:20

    • ๐Ÿท๏ธ Title: FF-Midj-Rise-v0564

    • ๐Ÿ–ผ๏ธ Resolution: 1024x1024

    • ๐Ÿงช Architecture: stable-diffusion-xl-v1-base/lora

    • ๐ŸŒ Network Dimensions:

      • Dim/Rank: 64.0

      • Alpha: 32.0

    • ๐Ÿ”Œ Module: networks.lora

    • ๐Ÿ“ˆ Stats:

      • UNet Weight (Avg. Magnitude): 2.6016

      • UNet Weight (Avg. Strength): 0.0074

      • Text Encoder (1) - Weight (Avg. Magnitude): 2.5694

      • Text Encoder (1) - Weight (Avg. Strength): 0.0091

      • Text Encoder (2) - Weight (Avg. Magnitude): 2.6260

      • Text Encoder (2) - Weight (Avg. Strength): 0.0071

    ๐Ÿ“‚ FF-Midj-Top-v0564-FA-TX.safetensors

    • ๐Ÿ“… Date: 2023-09-26T03:21:49

    • ๐Ÿท๏ธ Title: FF-Midj-Top-v0564-FA-TX

    • ๐Ÿ–ผ๏ธ Resolution: 1024x1024

    • ๐Ÿงช Architecture: stable-diffusion-xl-v1-base/lora

    • ๐ŸŒ Network Dimensions:

      • Dim/Rank: 64.0

      • Alpha: 64.0

    • ๐Ÿ”Œ Module: networks.lora_fa

    • ๐Ÿ“ˆ Stats:

      • Text Encoder (1) - Weight (Avg. Magnitude): 5.8341

      • Text Encoder (1) - Weight (Avg. Strength): 0.0191

      • Text Encoder (2) - Weight (Avg. Magnitude): 6.0269

      • Text Encoder (2) - Weight (Avg. Strength): 0.0153

    • โš ๏ธ Note: No UNet found in this LoRA.

    ๐ŸŒ FFusion.ai Contact Information

    Proudly maintained by Source Code Bulgaria Ltd & Black Swan Technologies.

    • ๐Ÿ“ง Email Us: [email protected] - For inquiries or support.

    • ๐ŸŒ Locations: Sofia | Istanbul | London

    Connect with Us:

    Our Websites:

    Description

    Update (11/02/23) : adding WEEK 3 FFUsion (ALL IN) LoRA of Hot, Rising, and New

    FAQ

    LORA
    SDXL 1.0
    by idle

    Details

    Downloads
    1,287
    Platform
    SeaArt
    Platform Status
    Available
    Created
    6/7/2024
    Updated
    11/2/2023
    Deleted
    -

    Files

    Available On (1 platform)

    Same model published on other platforms. May have additional downloads or version variants.