config.json

{
  "architectures": [
    "PixtralVisionModel"
  ],
  "attention_dropout": 0.0,
  "head_dim": 64,
  "hidden_act": "gelu",
  "hidden_size": 1024,
  "image_size": 1024,
  "intermediate_size": 4096,
  "is_composition": true,
  "model_type": "pixtral",
  "num_attention_heads": 16,
  "num_channels": 3,
  "num_hidden_layers": 24,
  "patch_size": 16,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.46.0.dev0"
}