Llm === LLM --- CrossAttentionGPT2 ~~~~~~~~~~~~~~~~~~ .. mermaid:: classDiagram class CrossAttentionGPT2 { +embed_tokens() Tensor +forward() Tensor } class Module <> Module Module <|-- CrossAttentionGPT2 : herits .. autosummary:: :toctree: generated :nosignatures: ~mfai.pytorch.models.llms.gpt2.CrossAttentionGPT2 GPT2 ~~~~ .. mermaid:: classDiagram class GPT2 { +download_weights_from_tf_ckpt() None +embed_tokens() Tensor +forward() Tensor +forward_vectors() Tensor +load_weights_from_dict() +reset_kv_cache() None } class Module <> Module Module <|-- GPT2 : herits .. autosummary:: :toctree: generated :nosignatures: ~mfai.pytorch.models.llms.gpt2.GPT2 Llama2 ~~~~~~ .. mermaid:: classDiagram class Llama2 { +embed_tokens() Tensor +forward() Tensor +forward_vectors() Tensor } class Module <> Module Module <|-- Llama2 : herits .. autosummary:: :toctree: generated :nosignatures: ~mfai.pytorch.models.llms.llama2.Llama2 Llama3 ~~~~~~ .. mermaid:: classDiagram class Llama3 { +embed_tokens() Tensor +forward() Tensor +forward_vectors() Tensor +reset_kv_cache() None } class Module <> Module Module <|-- Llama3 : herits .. autosummary:: :toctree: generated :nosignatures: ~mfai.pytorch.models.llms.llama3.Llama3 Qwen3_5 ~~~~~~~ .. mermaid:: classDiagram class Qwen3_5 { +compute_memory_size() float +create_mask() Tensor +download_weights_from_hf() None +forward() Tensor +generate_output_stream() Iterator +generate_text() str +load_weights_from_dict() None +reset_kv_cache() None } class Module <> Module Module <|-- Qwen3_5 : herits .. autosummary:: :toctree: generated :nosignatures: ~mfai.pytorch.models.llms.qwen3_5.Qwen3_5