# MiniCPM-o 4.5 two-GPU continuous-batching layout.
#
# Thinker owns GPU 0. Talker and Code2Wav share GPU 1, keeping text decoding
# isolated from codec/vocoder contention while preserving stage-2 batching.
pipeline: minicpmo_4_5
async_chunk: true

trust_remote_code: true
enable_prefix_caching: false

connectors:
  connector_of_shared_memory:
    name: SharedMemoryConnector
    extra:
      connector_get_sleep_s: 0.01
      connector_get_max_wait_first_chunk: 3000
      connector_get_max_wait: 300
      codec_chunk_frames: 25
      codec_left_context_frames: 3
      # HiFT capture buckets are derived from codec_chunk_frames and
      # codec_left_context_frames. (#5869)
      # A future initial_codec_chunk_frames shape will use lazy capture unless
      # HiFTGraphWrapper.derive_capture_bucket_size explicitly includes it.
      enable_hift_graph: true
      enable_cfm_graph: true
      cfm_max_graphs: 32

stages:
  - stage_id: 0
    max_num_seqs: 4
    gpu_memory_utilization: 0.9
    enforce_eager: false
    tensor_parallel_size: 1
    max_num_batched_tokens: 16384
    devices: "0"
    limit_mm_per_prompt:
      video: 1
    default_sampling_params:
      temperature: 0.0
      top_p: 1.0
      top_k: -1
      max_tokens: 2048
      seed: 42
      detokenize: true
      repetition_penalty: 1.1

  - stage_id: 1
    max_num_seqs: 4
    gpu_memory_utilization: 0.55
    enforce_eager: false
    max_num_batched_tokens: 8192
    devices: "1"
    output_connectors:
      to_stage_2: connector_of_shared_memory
    default_sampling_params:
      temperature: 0.8
      top_p: 0.8
      top_k: 100
      repetition_penalty: 1.02
      min_tokens: 50
      max_tokens: 4096
      seed: 42
      detokenize: false

  - stage_id: 2
    max_num_seqs: 4
    gpu_memory_utilization: 0.35
    enforce_eager: true
    enable_chunked_prefill: false
    max_num_batched_tokens: 65536
    max_model_len: 65536
    devices: "1"
    input_connectors:
      from_stage_1: connector_of_shared_memory
    default_sampling_params:
      temperature: 0.0
      top_p: 1.0
      top_k: -1
      max_tokens: 65536
      detokenize: true

platforms:
  npu:
    stages:
      - stage_id: 0
        max_num_batched_tokens: 8192
        compilation_config:
          cudagraph_mode: PIECEWISE
      - stage_id: 1
        max_num_batched_tokens: 8192
        compilation_config:
          cudagraph_mode: PIECEWISE
      - stage_id: 2
        # Stage 2 stays outer-eager and captures only exact-shape CFM DiT
        # estimator kernels through the NPU platform patch.
        additional_config:
          code2wav_enable_npu_graph: true
          code2wav_max_npu_graphs: 32