llm-multimodal 1.0.0

Multimodal processing for vision and other modalities
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

pub mod error;
pub mod media;
pub mod registry;
pub mod tracker;
pub mod types;
pub mod vision;

pub use error::{MediaConnectorError, MultiModalError, MultiModalResult};
pub use media::{ImageFetchConfig, MediaConnector, MediaConnectorConfig, MediaSource};
pub use registry::{ModelProcessorSpec, ModelRegistry};
pub use tracker::{AsyncMultiModalTracker, TrackerConfig, TrackerOutput};
pub use types::{
    ChatContentPart, ConversationSegment, ImageDetail, ImageFrame, ImageSize, ImageSource,
    Modality, MultiModalData, MultiModalInputs, MultiModalTensor, MultiModalUUIDs, MultiModalValue,
    PlaceholderHandle, PlaceholderMap, PlaceholderRange, PromptReplacement, TokenId, TrackedMedia,
    DEFAULT_PLACEHOLDERS,
};
// Re-export vision processing components
pub use vision::{
    ImagePreProcessor, ImageProcessorRegistry, LlavaNextProcessor, LlavaProcessor,
    PreProcessorConfig, PreprocessedImages, TransformError,
};