Melon-Bread/anima-pc98-gal-style-v0.2

Anima

Prompt

pc98gal_style, dithering, limited color palette, flat cel shading, thin linework, dynamic battle scene, young female mage with long teal hair in high ponytail, white and gold magical armor, wielding a glowing blue staff casting lightning spell, energy arcs and sparks around her, dark dramatic sky background, motion blur, action pose mid-cast, fantasy RPG battle scene

Negative Prompt

modern, 3d, realistic, photograph, blurry, low quality, deformed, extra digits, watermark, text, signature

Prompt

pc98gal_style, dithering, limited color palette, flat cel shading, thin linework, dramatic scene, young swordsman with long brown hair, red armor, holding a katana, mountain cliff, sunset sky, dramatic lighting, wind blowing hair, action pose

Negative Prompt

modern, 3d, realistic, photograph, blurry, low quality, deformed, extra digits, watermark, text, signature

Prompt

pc98gal_style, dithering, limited color palette, flat cel shading, science fiction scene, female captain with short red hair in military uniform standing at command console on spaceship bridge, viewports showing stars and nebula outside, multiple crew members in background at stations, blue holographic displays, retro sci-fi anime aesthetic, dramatic lighting from consoles

Negative Prompt

modern, 3d, realistic, photograph, blurry, low quality, deformed, extra digits, watermark, text, signature

Model description

pc98gal_style-v0.2 (Anima) — Changelog

What's new

• Trained on 87 PC-98 galgame images (up from 76 in earlier attempts — 11 new images from `harenochi`, `poisonneedle`, `ram`, `tsuioku`)
• All captions regenerated via ToriiGate-0.5 captioning script (`scripts/batch_caption_torii.py`) — natural-language descriptions prefixed with `pc98gal_style` trigger token
• 5 checkpoints available as separate safetensors (`-000002`, `-000004`, `-000006`, `-000008`, `-000010`) + final

Training config

• Base model: Anima base-v1.0 (DiT + Qwen3-0.6B + flow matching)
• LoRA rank: 32, alpha 32
• Optimizer: AdamW, lr 5e-5, weight_decay 0.1, betas [0.9, 0.99]
• Scheduler: cosine_with_restarts, 100 warmup steps
• Training: 87 images × 3 repeats × 10 epochs = 2610 steps, batch size 1
• Precision: bf16 throughout, gradient checkpointing on
• Resolution: 768×768
• Caption dropout: 10%
• No flip augmentation — PC-98 linework is directional
• Duration: ~2h17m on Radeon 8060S (Strix Halo, gfx1151, ROCm 7.13.99)

Results

• Epochs: 10
• Total steps: 2610
• Final loss: 0.123 (started ~0.13, converged smoothly)

Trigger token

`pc98gal_style` — must appear at the start of every prompt

Trigger words

You should use pc98gal_style to trigger the image generation.

Download model

Download them in the Files & versions tab.