Mistral3Tokenizer

[source]

Mistral3Tokenizer class

keras_hub.tokenizers.Mistral3Tokenizer(
    vocabulary=None, merges=None, split_pattern=None, control_tokens=None, **kwargs
)

Mistral3 Tekken tokenizer layer based on byte-level BPE.

Handles Mistral3's Tekken (tiktoken-style byte-level BPE) vocabulary. It is based on keras_hub.tokenizers.BytePairTokenizer, uses the Tekken pre-tokenization regex instead of the GPT-2/Llama3 pattern hardcoded in the base class, and registers the special image tokens ("[IMG]", "[IMG_BREAK]", "[IMG_END]") used by multimodal Mistral3/Pixtral models to mark image placeholder positions in a prompt.

Not a subclass of keras_hub.models.MistralTekkenTokenizer: that class builds its unsplittable_tokens list inline in __init__ before calling the BytePairTokenizer constructor, which doesn't leave a clean extension point for adding the vision tokens via inheritance.

The vocabulary and merges are usually produced from a tekken.json file by the Hugging Face conversion path; see keras_hub.src.utils.transformers.convert_mistral3.

If input is a batch of strings (rank > 0), the layer will output a tf.RaggedTensor where the last dimension of the output is ragged.

If input is a scalar string (rank == 0), the layer will output a dense tf.Tensor with static shape [None].

Arguments

  • vocabulary: A dict mapping token strings to integer ids, or a path to a vocabulary JSON file.
  • merges: A list of BPE merge rules, or a path to a merges file.
  • split_pattern: str, optional. The Tekken pre-tokenization regex. Defaults to the pattern shared by known Mistral3 checkpoints.
  • control_tokens: list of str, optional. Extra reserved control tokens (e.g. "[INST]") to register as unsplittable, in addition to the start/end/vision tokens. Defaults to None.

Examples

tokenizer = keras_hub.models.Mistral3Tokenizer.from_preset(
    "hf://mistralai/Mistral-Small-3.2-24B-Instruct-2506",
)
tokenizer("The quick brown fox jumped.")
tokenizer.detokenize(tokenizer("The quick brown fox jumped."))

[source]

from_preset method

Mistral3Tokenizer.from_preset(preset, config_file="tokenizer.json", **kwargs)

Instantiate a keras_hub.models.Tokenizer 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 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'

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

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

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 preset tokenizer.
tokenizer = keras_hub.tokenizer.Tokenizer.from_preset("bert_base_en")

# Tokenize some input.
tokenizer("The quick brown fox tripped.")

# Detokenize some input.
tokenizer.detokenize([5, 6, 7, 8, 9])
Preset Parameters Description
ministral_3_3b_base_2512_en 3.85B 3 billion parameter, 26-layer, pretrained Ministral 3 model with a Pixtral vision encoder for image input.
ministral_3_3b_instruct_2512_en 3.85B 3 billion parameter, 26-layer, instruction-tuned Ministral 3 model with a Pixtral vision encoder for image input.
ministral_3_3b_reasoning_2512_en 3.85B 3 billion parameter, 26-layer, reasoning-tuned Ministral 3 model with a Pixtral vision encoder for image input.
shieldstral_1.0_3b_en 3.85B 3 billion parameter, 26-layer Shieldstral model with a Pixtral vision encoder. The model checks text and image content against a safety policy written in natural language. The model returns one safety score for each input.
ministral_3_8b_base_2512_en 8.92B 8 billion parameter, 34-layer, pretrained Ministral 3 model with a Pixtral vision encoder for image input.
ministral_3_8b_instruct_2512_en 8.92B 8 billion parameter, 34-layer, instruction-tuned Ministral 3 model with a Pixtral vision encoder for image input.
ministral_3_8b_reasoning_2512_en 8.92B 8 billion parameter, 34-layer, reasoning-tuned Ministral 3 model with a Pixtral vision encoder for image input.
ministral_3_14b_base_2512_en 13.95B 14 billion parameter, 40-layer, pretrained Ministral 3 model with a Pixtral vision encoder for image input.
ministral_3_14b_instruct_2512_en 13.95B 14 billion parameter, 40-layer, instruction-tuned Ministral 3 model with a Pixtral vision encoder for image input.
ministral_3_14b_reasoning_2512_en 13.95B 14 billion parameter, 40-layer, reasoning-tuned Ministral 3 model with a Pixtral vision encoder for image input.
mistral_small_3.1_24b_base_2503_en 24.01B 24 billion parameter, 40-layer, pretrained Mistral3 model with a Pixtral vision encoder for image input.
mistral_small_3.1_24b_instruct_2503_en 24.01B 24 billion parameter, 40-layer, instruction-tuned Mistral3 model with a Pixtral vision encoder for image input.
mistral_small_3.2_24b_instruct_2506_en 24.01B 24 billion parameter, 40-layer, instruction-tuned Mistral3 model with a Pixtral vision encoder for image input. An updated version of mistral_small_3.1_24b_instruct_2503_en with improved instruction-following and reduced repetition.
devstral_small_2_24b_instruct_2512_en 24.01B 24 billion parameter, 40-layer, instruction-tuned Devstral Small 2 model for agentic software engineering, with a Pixtral vision encoder for image input.
magistral_small_2509_en 24.01B 24 billion parameter, 40-layer, reasoning-tuned Magistral Small model with a Pixtral vision encoder for image input.