Qwen3Backbone model

[source]

Qwen3Backbone class

keras_hub.models.Qwen3Backbone(
    vocabulary_size,
    num_layers,
    num_query_heads,
    num_key_value_heads,
    head_dim,
    hidden_dim,
    intermediate_dim,
    rope_max_wavelength=10000,
    rope_scaling_factor=1.0,
    layer_norm_epsilon=1e-06,
    dropout=0.0,
    tie_word_embeddings=True,
    sliding_window_size=32768,
    dtype=None,
    **kwargs
)

The Qwen3 Transformer core architecture with hyperparameters.

This network implements a Transformer-based decoder network, Qwen3, as described in the Qwen3 model architecture. It includes the embedding lookups and transformer layers.

The default constructor gives a fully customizable, randomly initialized Qwen3 model with any number of layers, heads, and embedding dimensions. To load preset architectures and weights, use the from_preset constructor.

Arguments

• vocabulary_size (int): The size of the token vocabulary.
• num_layers (int): The number of transformer layers.
• num_query_heads (int): The number of query attention heads for each transformer.
• hidden_dim (int): The size of the transformer encoding and pooling layers.
• intermediate_dim (int): The output dimension of the first Dense layer in a three-layer feedforward network for each transformer.
• num_key_value_heads (int): The number of key and value attention heads for each transformer.
• rope_max_wavelength (int, optional): The maximum angular wavelength of the sine/cosine curves, for rotary embeddings. Defaults to 10000.
• rope_scaling_factor (float, optional): The scaling factor for calculation of rotary embedding. Defaults to 1.0.
• layer_norm_epsilon (float, optional): Epsilon for the layer normalization layers in the transformer decoder. Defaults to 1e-6.
• dropout (float, optional): Dropout rate for attention and hidden layers. Defaults to 0.
• dtype: string or keras.mixed_precision.DTypePolicy. The dtype to use for model computations and weights. Note that some computations, such as softmax and layer normalization, will always be done at float32 precision regardless of dtype.
• tie_word_embeddings (bool, optional): Whether to tie input and output embeddings. Defaults to True.
• sliding_window_size (int, optional): Size of the sliding window for attention when enabled. Defaults to 32768.

Examples

input_data = {
    "token_ids": np.ones(shape=(1, 12), dtype="int32"),
    "padding_mask": np.array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0]]),
}

# Pretrained Qwen3 decoder.
model = keras_hub.models.Qwen3Backbone.from_preset("qwen32.5_0.5b_en")
model(input_data)

# Randomly initialized Qwen3 decoder with custom config.
model = keras_hub.models.Qwen3Backbone(
    vocabulary_size=10,
    hidden_dim=512,
    num_layers=2,
    num_query_heads=32,
    num_key_value_heads=8,
    intermediate_dim=1024,
    layer_norm_epsilon=1e-6,
    dtype="float32"
)
model(input_data)

[source]

from_preset method

Qwen3Backbone.from_preset(preset, load_weights=True, **kwargs)

Instantiate a keras_hub.models.Backbone from a model preset.

A preset is a directory of configs, weights and other file assets used to save and load a pre-trained model. The preset can be passed as a one of:

1. a built-in preset identifier like 'bert_base_en'
2. a Kaggle Models handle like 'kaggle://user/bert/keras/bert_base_en'
3. a Hugging Face handle like 'hf://user/bert_base_en'
4. a path to a local preset directory like './bert_base_en'

This constructor can be called in one of two ways. Either from the base class like keras_hub.models.Backbone.from_preset(), or from a model class like keras_hub.models.GemmaBackbone.from_preset(). If calling from the base class, the subclass of the returning object will be inferred from the config in the preset directory.

For any Backbone subclass, you can run cls.presets.keys() to list all built-in presets available on the class.

Arguments

• preset: string. A built-in preset identifier, a Kaggle Models handle, a Hugging Face handle, or a path to a local directory.
• load_weights: bool. If True, the weights will be loaded into the model architecture. If False, the weights will be randomly initialized.

Examples

# Load a Gemma backbone with pre-trained weights.
model = keras_hub.models.Backbone.from_preset(
    "gemma_2b_en",
)

# Load a Bert backbone with a pre-trained config and random weights.
model = keras_hub.models.Backbone.from_preset(
    "bert_base_en",
    load_weights=False,
)

token_embedding property

keras_hub.models.Qwen3Backbone.token_embedding

A keras.layers.Embedding instance for embedding token ids.

This layer embeds integer token ids to the hidden dim of the model.

[source]

enable_lora method

Qwen3Backbone.enable_lora(rank, target_layer_names=None)

Enable Lora on the backbone.

Calling this method will freeze all weights on the backbone, while enabling Lora on the query & value EinsumDense layers of the attention layers.

Arguments

• rank: The rank of the LoRA factorization.
• target_layer_names: A list of strings, the names of the layers to apply LoRA to. If None, this will be populated with the default LoRA layer names as returned by backbone.default_lora_layer_names().