![]() |
LlamaLib
v2.0.2
Cross-platform library for local LLMs
|
Core LLM functionality interface and base classes. More...
Go to the source code of this file.
Classes | |
| struct | LoraIdScale |
| Structure representing a LoRA adapter with ID and scale. More... | |
| struct | LoraIdScalePath |
| Structure representing a LoRA adapter with ID, scale, and file path. More... | |
| class | LLM |
| Abstract base class for Large Language Model operations. More... | |
| class | LLMLocal |
| Abstract class for local LLM operations with slot management. More... | |
| class | LLMProvider |
| Abstract class for LLM service providers. More... | |
| class | LLMProviderRegistry |
| Registry for managing LLM provider instances. More... | |
Functions | |
| void | ensure_error_handlers_initialized () |
| Ensures error handlers are properly initialized. | |
| bool | Has_GPU_Layers (const char *command) |
| Check if command has GPU layers (C API) | |
| void | LLM_Debug (int debug_level) |
| Set global debug level (C API) | |
| void | LLM_Logging_Callback (CharArrayFn callback) |
| Set global logging callback (C API) | |
| void | LLM_Logging_Stop () |
| Stop global logging (C API) | |
| void | LLM_Set_Completion_Parameters (LLM *llm, const char *params_json="{}") |
| Set completion parameters (C API) | |
| const char * | LLM_Get_Completion_Parameters (LLM *llm) |
| Get completion parameters (C API) | |
| void | LLM_Set_Grammar (LLM *llm, const char *grammar="") |
| Set grammar (C API) | |
| const char * | LLM_Get_Grammar (LLM *llm) |
| Get grammar (C API) | |
| const char * | LLM_Apply_Template (LLM *llm, const char *messages_as_json) |
| Apply chat template (C API) | |
| const char * | LLM_Tokenize (LLM *llm, const char *query) |
| Tokenize text (C API) | |
| const char * | LLM_Detokenize (LLM *llm, const char *tokens_as_json) |
| Detokenize tokens (C API) | |
| const char * | LLM_Embeddings (LLM *llm, const char *query) |
| Generate embeddings (C API) | |
| const char * | LLM_Completion (LLM *llm, const char *prompt, CharArrayFn callback=nullptr, int id_slot=-1, bool return_response_json=false) |
| Generate completion (C API) | |
| const char * | LLM_Save_Slot (LLMLocal *llm, int id_slot, const char *filepath) |
| Save slot state (C API) | |
| const char * | LLM_Load_Slot (LLMLocal *llm, int id_slot, const char *filepath) |
| Load slot state (C API) | |
| void | LLM_Cancel (LLMLocal *llm, int id_slot) |
| Cancel request (C API) | |
| bool | LLM_Lora_Weight (LLMProvider *llm, const char *loras_as_json) |
| Configure LoRA weights (C API) | |
| void | LLM_Enable_Reasoning (LLMProvider *llm, bool enable_reasoning) |
| Enable reasoning (C API) | |
| const char * | LLM_Lora_List (LLMProvider *llm) |
| List LoRA adapters (C API) | |
| void | LLM_Delete (LLMProvider *llm) |
| Delete LLM provider (C API) | |
| void | LLM_Start (LLMProvider *llm) |
| Start LLM service (C API) | |
| const bool | LLM_Started (LLMProvider *llm) |
| Check if service is started (C API) | |
| void | LLM_Stop (LLMProvider *llm) |
| Stop LLM service (C API) | |
| void | LLM_Start_Server (LLMProvider *llm, const char *host="0.0.0.0", int port=-1, const char *API_key="") |
| Start HTTP server (C API) | |
| void | LLM_Stop_Server (LLMProvider *llm) |
| Stop HTTP server (C API) | |
| void | LLM_Join_Service (LLMProvider *llm) |
| Wait for service to complete (C API) | |
| void | LLM_Join_Server (LLMProvider *llm) |
| Wait for server to complete (C API) | |
| void | LLM_Set_SSL (LLMProvider *llm, const char *SSL_cert, const char *SSL_key) |
| Set SSL configuration (C API) | |
| const int | LLM_Status_Code () |
| Get last operation status code (C API) | |
| const char * | LLM_Status_Message () |
| Get last operation status message (C API) | |
| const int | LLM_Embedding_Size (LLMProvider *llm) |
| Get embedding vector size (C API) | |
Core LLM functionality interface and base classes.
This file provides the foundational interface for Large Language Model operations, including completion, tokenization, embeddings, and LoRA (Low-Rank Adaptation) support.
Definition in file LLM.h.