Skip to main content

/

DocsCode

Python module

max.pipelines.architectures.idefics3_modulev3

Idefics3 vision-language architecture for multimodal text generation.

Idefics3Config​

class max.pipelines.architectures.idefics3_modulev3.Idefics3Config(*, devices, scale_factor, image_token_id, vision_config, text_config)

source

Bases: ArchConfigWithKVCache

Configuration for Idefics3 models (ModuleV3).

Parameters:

calculate_max_seq_len()​

static calculate_max_seq_len(pipeline_config, huggingface_config)

source

Calculate maximum sequence length for Idefics3.

Parameters:

Return type:

int

construct_kv_params()​

static construct_kv_params(huggingface_config, pipeline_config, devices, kv_cache_config, cache_dtype)

source

Get KV cache parameters for the language model.

Parameters:

Return type:

KVCacheParams

devices​

devices: list[DeviceRef]

source

Devices that the Idefics3 model is parallelized over.

finalize()​

finalize(huggingface_config, llm_state_dict, return_logits, return_hidden_states=ReturnHiddenStates.NONE, norm_method='rms_norm')

source

Finalize the Idefics3Config with state_dict-dependent fields.

Parameters:

Return type:

None

get_kv_params()​

get_kv_params()

source

Returns the KV cache parameters from the embedded text config.

Return type:

KVCacheParams

get_max_seq_len()​

get_max_seq_len()

source

Returns the maximum sequence length from the embedded text config.

Return type:

int

get_num_layers()​

static get_num_layers(huggingface_config)

source

Get number of layers in the language model.

Parameters:

huggingface_config (AutoConfig)

Return type:

int

image_seq_len​

property image_seq_len: int

source

Calculate the number of image tokens after connector processing.

image_token_id​

image_token_id: int

source

Token ID used to represent image tokens in the text sequence.

initialize()​

classmethod initialize(pipeline_config, model_config=None)

source

Initializes an Idefics3Config instance from pipeline configuration.

Parameters:

Return type:

Self

scale_factor​

scale_factor: int

source

Scale factor for pixel shuffle operation in the connector.

text_config​

text_config: Llama3Config

source

Text model configuration (Llama3-based).

vision_config​

vision_config: Idefics3VisionConfig

source

Vision encoder configuration (SigLIP-based).

Idefics3Model​

class max.pipelines.architectures.idefics3_modulev3.Idefics3Model(pipeline_config, session, devices, kv_cache_config, weights, adapter=None, return_logits=ReturnLogits.LAST_TOKEN)

source

Bases: PipelineModelWithKVCache[TextAndVisionContext]

An Idefics3 pipeline model using the ModuleV3 API.

Parameters:

calculate_max_seq_len()​

static calculate_max_seq_len(pipeline_config, huggingface_config)

source

Calculates the optimal max sequence length for the model.

Models are expected to implement this method. The following example shows how to implement it for a Mistral model:

class MistralModel(PipelineModel):
    @classmethod
    def calculate_max_seq_len(cls, pipeline_config, huggingface_config) -> int:
        try:
            return upper_bounded_default(
                upper_bound=huggingface_config.max_seq_len,
                default=pipeline_config.model.max_length,
            )
        except ValueError as e:
            raise ValueError(
                "Unable to infer max_length for Mistral, the provided "
                f"max_length ({pipeline_config.model.max_length}) exceeds the "
                f"model's max_seq_len ({huggingface_config.max_seq_len})."
            ) from e

Parameters:

  • pipeline_config (PipelineConfig) – Configuration for the pipeline.
  • huggingface_config (AutoConfig) – Hugging Face model configuration.

Returns:

The maximum sequence length to use.

Return type:

int

execute()​

execute(model_inputs)

source

Execute the Idefics3 model.

Parameters:

model_inputs (ModelInputs)

Return type:

ModelOutputs

get_kv_params()​

classmethod get_kv_params(huggingface_config, pipeline_config, devices, kv_cache_config, cache_dtype)

source

Returns the KV cache params for the pipeline model.

Parameters:

Return type:

KVCacheParams

language_model​

language_model: Callable[..., Any]

source

The compiled language model.

load_model()​

load_model()

source

Compile vision and language models using the V3 API.

Returns:

A tuple of (compiled_vision_model, compiled_language_model).

Return type:

tuple[Callable[[…], Any], Callable[[…], Any]]

prepare_initial_token_inputs()​

prepare_initial_token_inputs(replica_batches, kv_cache_inputs=None, return_n_logits=1)

source

Prepare the initial inputs for the first execution pass.

Parameters:

Return type:

ModelInputs

prepare_next_token_inputs()​

prepare_next_token_inputs(next_tokens, prev_model_inputs)

source

Prepares the secondary inputs to be passed to execute().

While prepare_initial_token_inputs is responsible for managing the initial inputs. This function is responsible for updating the inputs, for each step in a multi-step execution pattern.

Parameters:

Return type:

Idefics3Inputs

vision_model​

vision_model: Callable[..., Any]

source

The compiled vision model.

Models

Cookbook

Puzzles

Blog

Community

Contact