job: extension
config:
  name: n1sr1n4
  process:
  - type: diffusion_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: 500
      max_step_saves_to_keep: 4
      save_format: diffusers
      push_to_hub: false
    datasets:
    - folder_path: /app/ai-toolkit/datasets/n1sr1n4
      mask_path: null
      mask_min_value: 0.1
      default_caption: ''
      caption_ext: txt
      caption_dropout_rate: 0.05
      cache_latents_to_disk: true
      is_reg: false
      network_weight: 1
      resolution:
      - 512
      - 768
      controls: []
      shrink_video_to_frames: true
      num_frames: 121
      flip_x: false
      flip_y: false
      num_repeats: 1
      do_i2v: false
      do_audio: true
      fps: 24
      audio_normalize: true
    train:
      batch_size: 1
      bypass_guidance_embedding: false
      steps: 6500
      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: true
      lr: 0.0001
      ema_config:
        use_ema: false
        ema_decay: 0.99
      skip_first_sample: false
      force_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
      loss_type: mse
    logging:
      log_every: 1
      use_ui_logger: true
    model:
      name_or_path: Lightricks/LTX-2
      quantize: false
      qtype: qfloat8
      quantize_te: true
      qtype_te: uint4
      arch: ltx2
      low_vram: true
      model_kwargs: {}
      layer_offloading: false
      layer_offloading_text_encoder_percent: 1
      layer_offloading_transformer_percent: 1
    sample:
      sampler: flowmatch
      sample_every: 500
      width: 768
      height: 768
      samples:
      - prompt: A cinematic medium shot of n1sr1n4 standing inside a metallic elevator.
          She looks tired, wearing a professional black blazer over a white patterned
          top and a blue lanyard around her neck. She holds a coffee cup loosely in
          her hand. The elevator doors start to close, but she looks up, makes eye
          contact with the viewer, and says softly, "It is going to be a long day."
          The lighting is cool and artificial, reflecting off the metal walls.
        ctrl_img: /app/ai-toolkit/data/images/fc426fd4-05a1-4093-9115-bcbd2462a2db.jpg
      - prompt: A close-up shot of n1sr1n4 outdoors with sunlight hitting her face.
          She takes a drag from her cigarette and exhales smoke slowly. She is wearing
          her signature blue patterned shirt. The camera focuses on her face as she
          notices the recording, smiles slightly, and says playfully, "Are you really
          filming me right now?" The background is a blur of city traffic and trees.
      - prompt: A handheld vlog-style shot of n1sr1n4 sitting in a bright coffee shop.
          She is holding her iced coffee cup. She looks directly into the camera lens,
          tilts her head slightly, and says, "I think this is the best coffee I have
          ever had." After speaking, she smiles and takes a sip from the straw. The
          background is slightly blurred with people walking by.
      neg: ''
      seed: 42
      walk_seed: true
      guidance_scale: 4
      sample_steps: 25
      num_frames: 121
      fps: 24
meta:
  name: n1sr1n4
  version: '1.0'