# MammothModa2 AR→DiT (text-to-image).

async_chunk: false
pipeline: mammoth_moda2

stages:
  - stage_id: 0
    devices: "0"
    max_num_seqs: 100
    max_model_len: 8192
    gpu_memory_utilization: 0.5
    enforce_eager: true
    trust_remote_code: true
    enable_prefix_caching: false

  - stage_id: 1
    devices: "0"
    max_num_seqs: 8
    request_batch_max_wait_ms: 20
    gpu_memory_utilization: 0.3
    enforce_eager: true
    trust_remote_code: true
    enable_prefix_caching: false
    # Optional Cache-DiT acceleration for the DiT stage (~2x faster denoise,
    # training-free). Uncomment to enable; guidance=1.0 requests bypass it.
    # cache_backend: cache_dit
    # cache_config:
    #   Fn_compute_blocks: 1
    #   Bn_compute_blocks: 0
    #   max_warmup_steps: 4
    #   residual_diff_threshold: 0.24
    #   max_continuous_cached_steps: 3
    # enable_cache_dit_summary: true