experimental_experiment.torch_models.llm_model_helper

experimental_experiment.torch_models.llm_model_helper.get_ai21_jamba_15_mini(inputs_as_dict: bool = False, **kwargs) → Tuple[Any, Tuple[Any, ...] | Dict[str, Any]][source]

Gets a non initialized model.

Parameters:

  • inputs_as_dict – returns dummy inputs as a dictionary or not

  • kwargs – to overwrite the configuration, example num_hidden_layers=1

Returns:

model, inputs

See ai21labs/AI21-Jamba-1.5-Mini/config.json.

experimental_experiment.torch_models.llm_model_helper.get_all_mini_ml_l6_v1(inputs_as_dict: bool = False, **kwargs) → Tuple[Any, Tuple[Any, ...] | Dict[str, Any]][source]

Gets a non initialized model.

Parameters:

  • inputs_as_dict – returns dummy inputs as a dictionary or not

  • kwargs – to overwrite the configuration, example num_hidden_layers=1

Returns:

model, inputs

See all-MiniLM-L6-v1.

experimental_experiment.torch_models.llm_model_helper.get_falcon_mamba_7b(inputs_as_dict: bool = False, **kwargs) → Tuple[Any, Tuple[Any, ...] | Dict[str, Any]][source]

Gets a non initialized model.

Parameters:

  • inputs_as_dict – returns dummy inputs as a dictionary or not

  • kwargs – to overwrite the configuration, example num_hidden_layers=1

Returns:

model, inputs

See flacon-mamba-7b/config.json.

experimental_experiment.torch_models.llm_model_helper.get_llama_32_9b_vision(inputs_as_dict: bool = False, **kwargs) → Tuple[Any, Tuple[Any, ...] | Dict[str, Any]][source]

Gets a non initialized model.

Parameters:

  • inputs_as_dict – returns dummy inputs as a dictionary or not

  • kwargs – to overwrite the configuration, example num_hidden_layers=1

Returns:

model, inputs

See MLlama.

experimental_experiment.torch_models.llm_model_helper.get_phi_35_mini_instruct(inputs_as_dict: bool = False, batch: int = 1, **kwargs) → Tuple[Any, Tuple[Any, ...] | Dict[str, Any]][source]

Gets a non initialized model.

Parameters:

  • inputs_as_dict – returns dummy inputs as a dictionary or not

  • batch – batch size

  • kwargs – to overwrite the configuration, example num_hidden_layers=1

Returns:

model, inputs

See Phi-3.5-mini-instruct/config.json.

experimental_experiment.torch_models.llm_model_helper.get_phi_3_vision_128k_instruct(inputs_as_dict: bool = False, **kwargs) → Tuple[Any, Tuple[Any, ...] | Dict[str, Any]][source]

Gets a non initialized model.

Parameters:

  • inputs_as_dict – returns dummy inputs as a dictionary or not

  • kwargs – to overwrite the configuration, example num_hidden_layers=1

Returns:

model, inputs

See Phi-3-vision-128k-instruct/config.json.

experimental_experiment.torch_models.llm_model_helper.get_smollm_1_7b(inputs_as_dict: bool = False, **kwargs) → Tuple[Any, Tuple[Any, ...] | Dict[str, Any]][source]

Gets a non initialized model.

Parameters:

  • inputs_as_dict – returns dummy inputs as a dictionary or not

  • kwargs – to overwrite the configuration, example num_hidden_layers=1

Returns:

model, inputs

See SmolLM-1.7B.