job: extension
config:
  name: mindexpander_qwen_v1
  process:
  - type: ui_trainer
    training_folder: /app/ai-toolkit/output
    sqlite_db_path: /app/ai-toolkit/aitk_db.db
    device: cuda
    trigger_word: null
    performance_log_every: 10
    network:
      type: lora
      linear: 32
      linear_alpha: 32
      conv: 16
      conv_alpha: 16
      lokr_full_rank: true
      lokr_factor: -1
      network_kwargs:
        ignore_if_contains: []
    save:
      dtype: bf16
      save_every: 250
      max_step_saves_to_keep: 4
      save_format: diffusers
      push_to_hub: false
    datasets:
    - folder_path: /app/ai-toolkit/datasets/mind
      control_path: null
      mask_path: null
      mask_min_value: 0.1
      default_caption: ''
      caption_ext: txt
      caption_dropout_rate: 0.05
      cache_latents_to_disk: false
      is_reg: false
      network_weight: 1
      resolution:
      - 512
      - 768
      - 1024
      controls: []
      shrink_video_to_frames: true
      num_frames: 1
      do_i2v: true
      flip_x: false
      flip_y: false
    train:
      batch_size: 1
      bypass_guidance_embedding: false
      steps: 3000
      gradient_accumulation: 1
      train_unet: true
      train_text_encoder: false
      gradient_checkpointing: true
      noise_scheduler: flowmatch
      optimizer: adamw8bit
      timestep_type: weighted
      content_or_style: balanced
      optimizer_params:
        weight_decay: 0.0001
      unload_text_encoder: false
      cache_text_embeddings: false
      lr: 0.0001
      ema_config:
        use_ema: false
        ema_decay: 0.99
      skip_first_sample: false
      disable_sampling: false
      dtype: bf16
      diff_output_preservation: false
      diff_output_preservation_multiplier: 1
      diff_output_preservation_class: person
      switch_boundary_every: 1
    model:
      name_or_path: Qwen/Qwen-Image
      quantize: true
      qtype: qfloat8
      quantize_te: true
      qtype_te: qfloat8
      arch: qwen_image
      low_vram: true
      model_kwargs: {}
    sample:
      sampler: flowmatch
      sample_every: 250
      width: 1024
      height: 1024
      samples:
      - prompt: photorealistic portrait of a person wearing the m1nd3xpand3r mask
          (third-eye symbol, orange/blue markings) seated on a dim night metro; neon
          city lights streak outside the window at camera-left; shallow depth of field,
          35mm lens, soft fluorescent flicker, window reflections; calm, moody, cinematic
          lighting; sharp focus, high detail, filmic color grade
      - prompt: photorealistic close-up of a person wearing the m1nd3xpand3r mask
          in a narrow neon-lit alley at night; light rain with beads and drips on
          the mask; shallow depth of field, 50mm lens; slow cinematic vibe, cool neon
          rimlight with warm shop-sign fill; steam and puddle reflections; sharp focus,
          high detail, cinematic contrast
      neg: ''
      seed: 42
      walk_seed: true
      guidance_scale: 4
      sample_steps: 25
      num_frames: 1
      fps: 1
meta:
  name: mindexpander_qwen_v1
  version: '1.0'