{
  "architectures": [
    "HfMoondream"
  ],
  "auto_map": {
    "AutoConfig": "hf_moondream.HfConfig",
    "AutoModelForCausalLM": "hf_moondream.HfMoondream"
  },
  "config": {
    "skills": [
      "query",
      "caption",
      "detect",
      "point"
    ]
  },
  "model_type": "moondream3",
  "torch_dtype": "bfloat16",
  "transformers_version": "4.51.1",
  "text": {
    "dim": 2048,
    "ff_dim": 8192,
    "n_layers": 24,
    "vocab_size": 51200,
    "max_context": 4096,
    "n_heads": 32,
    "n_kv_heads": 32,
    "prefix_attn": 730,
    "group_size": null,
    "moe": {
      "num_experts": 64,
      "start_layer": 4,
      "experts_per_token": 8,
      "expert_inner_dim": 1024
    }
  },
  "vision": {
    "enc_dim": 1152,
    "enc_patch_size": 14,
    "enc_n_layers": 27,
    "enc_ff_dim": 4304,
    "enc_n_heads": 16,
    "proj_out_dim": 2048,
    "crop_size": 378,
    "in_channels": 3,
    "max_crops": 12,
    "overlap_margin": 4,
    "proj_inner_dim": 8192
  },
  "region": {
    "dim": 2048,
    "coord_feat_dim": 256,
    "coord_out_dim": 1024,
    "size_feat_dim": 512,
    "size_out_dim": 2048,
    "group_size": null
  },
  "dtype": "bfloat16"
}