config.json

{
  "architectures": [
    "PenguinVLQwen3ForCausalLM"
  ],
  "auto_map": {
    "AutoConfig": "configuration_penguinvl.PenguinVLQwen3Config",
    "AutoModelForCausalLM": "modeling_penguinvl_qwen3.PenguinVLQwen3ForCausalLM"
  },
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "image_aspect_ratio": "square",
  "image_token_index": 151669,
  "initializer_range": 0.02,
  "intermediate_size": 12288,
  "loss_reduction_scope": "batch",
  "max_frames": 180,
  "max_position_embeddings": 40960,
  "max_window_layers": 36,
  "model_type": "penguinvl_qwen3",
  "num_attention_heads": 32,
  "num_hidden_layers": 36,
  "num_key_value_heads": 8,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000,
  "sliding_window": null,
  "tie_word_embeddings": false,
  "tokenizer_model_max_length": 32768,
  "tokenizer_padding_side": "right",
  "torch_dtype": "bfloat16",
  "transformers_version": "4.51.3",
  "use_cache": false,
  "use_sliding_window": false,
  "vision_encoder": "tencent/Penguin-Encoder",
  "vision_hidden_size": 1024,
  "vision_projector_type": "mlp2x_gelu",
  "vocab_size": 151936,
  "vision_encoder_config": {
    "head_dim": 128,
    "hidden_act": "silu",
    "hidden_size": 1024,
    "initializer_range": 0.02,
    "intermediate_size": 3072,
    "layer_norm_eps": 1e-06,
    "max_window_layers": 28,
    "num_attention_heads": 16,
    "num_channels": 3,
    "num_hidden_layers": 28,
    "num_key_value_heads": 8,
    "patch_size": 14,
    "rms_norm_eps": 1e-06,
    "rope_scaling": null,
    "rope_theta": 1000000,
    "sliding_window": null,
    "torch_dtype": "bfloat16"
  }
}