Anima/Rainbow - anima_rainbow.ft2.2

Anima/Rainbow

CHECKPOINT
Original


Updated:

Anima/Rainbow by Nano on Tensor.Art
Anima/Rainbow by Nano on Tensor.Art
Anima/Rainbow by Nano on Tensor.Art
Anima/Rainbow by Nano on Tensor.Art
Anima/Rainbow by Nano on Tensor.Art
Anima/Rainbow by Nano on Tensor.Art
Anima/Rainbow by Nano on Tensor.Art
Anima/Rainbow by Nano on Tensor.Art
Anima/Rainbow by Nano on Tensor.Art
Anima/Rainbow by Nano on Tensor.Art

This model aims to improve the base style generation capabilities of Anima base, enabling it to produce better results while guiding the output towards a cute anime art style.

I will do my best to maintain updates for this project, but ultimately it still depends on whether my GPU computing resources allow it.

Recommended Settings:

  • Sampler: er_sde simple

  • Recommended Resolution: 1024 × 1536

  • Prompt Guidance / Relevance: 4

  • Sampling Steps: 25–35

Recommended Positive Prompt: masterpiece,best quality,score_7,very aesthetic

Recommended Negative Prompt: worst quality,low quality,score_1,score_2,score_3,blurry,jpeg artifacts,sepia,muscular female

Please comply with all applicable local laws and regulations when using AI image generation models. Improper use may result in legal risks or other unintended consequences.

The actual usage of this model is determined solely by the user. The model creator cannot control or intervene in specific applications or generated outputs. Users are fully responsible for ensuring the legality, compliance, and potential impact of the generated content. Any consequences arising from the use of this model or its outputs shall be borne entirely by the user.

When using AI image generation models, please respect the rights of others and follow community guidelines. Avoid generating or distributing illegal, harmful, or inappropriate content, and help maintain a safe and healthy creative environment.

Version Detail

Anima
anima_rainbow.ft2.2kh: This version has increased the effective training data volume by 300%. It was trained on an RTX 5090 for approximately 90 hours. Compared to ft1, it improves the randomness of generation results, theoretically capable of producing more generalized outputs with softer colors. The trade-off is that, compared to ft1, it may be unable to maintain the fixed yet high-quality results; therefore, it is recommended to choose according to your needs.

Project Permissions

Reprinting is strictly prohibited

    Use Permissions

  • Use in TENSOR Online

  • As a online training base model on TENSOR

  • Use without crediting me

  • Share merges of this model

  • Use different permissions on merges

    Commercial Use

  • Sell generated contents

  • Use on generation services

  • Sell this model or merges

Related Posts