Functions — github.com/abetlen/llama-cpp-python

hub / github.com/abetlen/llama-cpp-python / functions

Functions1,336 in github.com/abetlen/llama-cpp-python

⨍ Functions 1,336◇ Types & classes 256↳ Endpoints 95

↓ 2 callersMethodrows_for_capacity

(self, offset: int, capacity: int)

examples/server/server.py:3886

↓ 2 callersMethodsave

( self, tokens: Sequence[int], state_bytes: np.ndarray, prompt_logits: Optiona

examples/server/server.py:4158

↓ 2 callersMethodset_batch

(self, batch: Sequence[int], n_past: int, logits_all: bool)

llama_cpp/_internals.py:505

↓ 2 callersMethodset_target_processing_enabled

(self, enabled: bool)

examples/server/server.py:1398

↓ 2 callersFunctionstream_sse_chunks

( formatter: OpenAIFormatter, http_request: Request, stream: CompletionStream,

examples/server/server.py:15830

↓ 2 callersMethodsubmit

( self, payload: CreateCompletionRequest, )

examples/server/server.py:15630

↓ 2 callersMethodto_chat_template_tool

(self)

examples/server/server.py:2951

↓ 2 callersMethodto_template_function_call

(self)

examples/server/server.py:2810

↓ 2 callersMethodto_template_json_schema

(self)

examples/server/server.py:3053

↓ 2 callersMethodto_template_tool

(self)

examples/server/server.py:2798

↓ 2 callersMethodtoken_bytes_with_prev_bytes

( self, prev_tokens: Sequence[int], prev_text_bytes: Union[bytes, bytearray],

examples/server/server.py:12070

↓ 2 callersMethodtoken_nl

Return the newline token.

llama_cpp/llama.py:2272

↓ 2 callersMethodtokenize

(self, text: bytes, add_bos: bool, special: bool)

llama_cpp/_internals.py:168

↓ 2 callersMethoduse_antiprompt

(self)

examples/low_level_api/low_level_api_chat_cpp.py:334

↓ 1 callersMethod__call__

( self, *, messages: List[llama_types.ChatCompletionRequestMessage], **kwargs:

llama_cpp/llama_chat_format.py:187

↓ 1 callersMethod__init__

Load a llama.cpp model from `model_path`. Examples: Basic usage >>> import llama_cpp >>> model = llama_c

llama_cpp/llama.py:60

↓ 1 callersMethod__init__

(self, capacity_bytes: int = (2 << 30))

llama_cpp/llama_cache.py:52

↓ 1 callersMethod_advance_direct_stream_state

(self, text: str)

examples/server/server.py:5836

↓ 1 callersMethod_advance_direct_tool_call_state

(self, text: str)

examples/server/server.py:5585

↓ 1 callersMethod_advance_json_scanner

( item_state: Dict[str, Any], text: str, *, schema_type: Optional[str], )

examples/server/server.py:5419

↓ 1 callersMethod_allocate_homogeneous_recurrent_draft_tokens

( self, sources: Sequence[Union[CompletionRequest, Completion]], token_count: int,

examples/server/server.py:13683

↓ 1 callersMethod_allocate_pending_tokens_for_model

( self, ordered_sequences: Sequence[Union[CompletionRequest, Completion]], )

examples/server/server.py:13627

↓ 1 callersMethod_allocate_recurrent_pending_only_tokens

( self, sources: Sequence[Union[CompletionRequest, Completion]], capacity: int, )

examples/server/server.py:13706

↓ 1 callersMethod_apply_message_delta

(message: Dict[str, Any], delta: Dict[str, Any])

examples/server/server.py:7485

↓ 1 callersFunction_assert_exact_cached_prompt_reuse_matches_fresh

( model_path, *, is_recurrent: bool, is_hybrid: bool, )

tests/test_llama.py:409

↓ 1 callersFunction_assert_loaded_exact_cached_prompt_reuse_matches_fresh

( model_path, *, is_recurrent: bool, is_hybrid: bool, )

tests/test_llama.py:449

↓ 1 callersFunction_assert_ram_cache_exact_prompt_hit_matches_fresh

( model_path, *, is_recurrent: bool, is_hybrid: bool, )

tests/test_llama.py:486

↓ 1 callersFunction_assert_shorter_prompt_prefix_reuse_matches_fresh

( model_path, *, is_recurrent: bool, is_hybrid: bool, )

tests/test_llama.py:541

↓ 1 callersMethod_build_chat_formatter

(self)

examples/server/server.py:11912

↓ 1 callersMethod_build_completion_choice

( self, request: CompletionRequest, completion: Completion, )

examples/server/server.py:10188

↓ 1 callersMethod_build_prompt_plan_locked

( self, *, messages: List[ChatCompletionRequestMessage], media_inputs: List[Me

examples/server/server.py:10998

↓ 1 callersMethod_build_sampled_batch_plan

( self, updates: Sequence["MTPDraftProvider.SampledBatchUpdate"], /, )

examples/server/server.py:2039

↓ 1 callersMethod_cached_repo_file_list

(self)

examples/server/server.py:3189

↓ 1 callersMethod_cached_stream_plan

( cls, response_schema: Optional[Dict[str, Any]], )

examples/server/server.py:5175

↓ 1 callersMethod_cached_tool_schema_map

( cls, tools: Optional[List[ChatTemplateTool]], )

examples/server/server.py:5210

↓ 1 callersMethod_chat_message_from_completion_choice

( completion_id: str, choice: CompletionChoice, tool_name: Optional[str], )

examples/server/server.py:9726

↓ 1 callersMethod_coerce_tool_arguments

( self, tool_name: str, arguments: Dict[str, str], *, partial: bool,

examples/server/server.py:7076

↓ 1 callersMethod_compile_iterator_pattern

(pattern: str)

examples/server/server.py:4763

↓ 1 callersMethod_compile_segment_message_plan

( cls, schema: Dict[str, Any], )

examples/server/server.py:4976

↓ 1 callersMethod_compile_stream_plan

( cls, response_schema: Optional[Dict[str, Any]], )

examples/server/server.py:5161

↓ 1 callersMethod_compile_tagged_message_plan

( cls, schema: Dict[str, Any], )

examples/server/server.py:5045

↓ 1 callersMethod_compile_word_capture_pattern

( cls, pattern: str, )

examples/server/server.py:4853

↓ 1 callersFunction_contains_dict_type

(annotation: Type[Any] | None)

llama_cpp/server/cli.py:44

↓ 1 callersFunction_contains_list_type

(annotation: Type[Any] | None)

llama_cpp/server/cli.py:33

↓ 1 callersMethod_contains_partial_json_value

(cls, value: Any)

examples/server/server.py:7302

↓ 1 callersMethod_content_part_for_template

( part: Any, media_marker: Optional[str], )

examples/server/server.py:3592

↓ 1 callersMethod_convert_content_part_for_template

( part: Any, media_marker: str, )

llama_cpp/llama_chat_format.py:3378

↓ 1 callersMethod_convert_message_for_template

( cls, message: llama_types.ChatCompletionRequestMessage, media_marker: str, )

llama_cpp/llama_chat_format.py:3363

↓ 1 callersFunction_convert_text_completion_chunks_to_chat

( chunks: Iterator[llama_types.CreateCompletionStreamResponse], )

llama_cpp/llama_chat_format.py:361

↓ 1 callersFunction_convert_text_completion_to_chat

( completion: llama_types.Completion, )

llama_cpp/llama_chat_format.py:335

↓ 1 callersMethod_create_bitmap_from_bytes

Create mtmd_bitmap from image bytes.

llama_cpp/llama_chat_format.py:2833

↓ 1 callersMethod_create_bitmap_from_bytes

(self, image_bytes: bytes)

llama_cpp/llama_chat_format.py:3327

↓ 1 callersMethod_create_completion

( self, prompt: Union[str, List[int]], suffix: Optional[str] = None, max_token

llama_cpp/llama.py:1188

↓ 1 callersMethod_create_loaded_media

( self, media: MediaInput, media_bytes: bytes, )

examples/server/server.py:10727

↓ 1 callersMethod_create_loaded_video_media

( self, media: MediaInput, media_bytes: bytes, key: str, )

examples/server/server.py:10776

↓ 1 callersMethod_decode_batch_for_mtp_heads

( self, start_pos_by_seq: Dict[int, int], )

examples/server/server.py:1485

↓ 1 callersMethod_decode_sampled_context_rows

( self, context_rows: Sequence["MTPDraftProvider.SampledContextRow"], h_tgt_rows: np.n

examples/server/server.py:2103

↓ 1 callersMethod_decode_sampled_pending_rows

( self, plan: "MTPDraftProvider.SampledBatchPlan", h_tgt_rows: np.ndarray, /,

examples/server/server.py:2137

↓ 1 callersMethod_draft_chain_heads

( self, *, seq_id: int, first_pos: int, token: int, n_predict:

examples/server/server.py:1657

↓ 1 callersMethod_embeddings_from_pointer

(self, output: Any, n_tokens: int)

examples/server/server.py:10848

↓ 1 callersMethod_encode_media_batch

( self, media_chunks: Sequence["MTMDProcessor.MediaChunk"], start_index: int, )

examples/server/server.py:10872

↓ 1 callersMethod_encode_media_chunks

( self, media_chunks: Sequence["MTMDProcessor.MediaChunk"], )

examples/server/server.py:10921

↓ 1 callersMethod_ensure_message_stream_item

( self, state: "OpenAIFormatter.ResponsesStream", )

examples/server/server.py:9234

↓ 1 callersMethod_ensure_reasoning_stream_item

( self, state: "OpenAIFormatter.ResponsesStream", )

examples/server/server.py:9203

↓ 1 callersMethod_ensure_sample_logits_buffer

(self, size: int)

examples/server/server.py:10341

↓ 1 callersMethod_ensure_tool_stream_item

( self, state: "OpenAIFormatter.ResponsesStream", *, tool_call_index: int,

examples/server/server.py:9266

↓ 1 callersFunction_eval_alternate_same_length_prompt

(model, tokens, expected_next_token)

tests/test_llama.py:385

↓ 1 callersMethod_extend_sampled_draft_states

( self, active: List["MTPDraftProvider.SampledDraftState"], results: List[np.ndarray],

examples/server/server.py:2239

↓ 1 callersFunction_format_add_colon_space_single

Format the prompt with the add-colon-space-single style.

llama_cpp/llama_chat_format.py:959

↓ 1 callersFunction_format_chatglm3

Format the prompt with the chatglm3 style.

llama_cpp/llama_chat_format.py:985

↓ 1 callersFunction_format_llama2

Format the prompt with the llama2 style.

llama_cpp/llama_chat_format.py:902

↓ 1 callersMethod_format_prometheus_value

(value: Union[int, float])

examples/server/server.py:15012

↓ 1 callersFunction_get_base_type

(annotation: Type[Any])

llama_cpp/server/cli.py:11

↓ 1 callersMethod_get_chat_template

(self, llama_model: llama.Llama)

llama_cpp/llama_chat_format.py:3344

↓ 1 callersMethod_get_template_messages

( self, messages: List[llama_types.ChatCompletionRequestMessage], media_marker: str,

llama_cpp/llama_chat_format.py:3352

↓ 1 callersFunction_grammar_for_json_schema

( schema: str, verbose: bool = False, fallback_to_json: bool = True )

llama_cpp/llama_chat_format.py:1006

↓ 1 callersMethod_grammar_for_response_format

( response_format: Optional[ChatTemplateResponseFormat], )

examples/server/server.py:8402

↓ 1 callersMethod_has_text_tools

(self)

examples/server/server.py:5393

↓ 1 callersMethod_init_mtmd_context

Initialize mtmd context with the llama model.

llama_cpp/llama_chat_format.py:2788

↓ 1 callersMethod_init_mtmd_context

(self, llama_model: llama.Llama)

llama_cpp/llama_chat_format.py:3285

↓ 1 callersMethod_init_samplers

(self)

examples/server/server.py:1336

↓ 1 callersMethod_instructions_role

(self)

examples/server/server.py:7929

↓ 1 callersMethod_load_cached_media_chunk

(self, media_chunk: "MTMDProcessor.MediaChunk")

examples/server/server.py:10855

↓ 1 callersMethod_load_entries

(self)

examples/server/server.py:12468

↓ 1 callersMethod_load_image

(image_url: str)

llama_cpp/llama_chat_format.py:3125

↓ 1 callersMethod_load_image

(image_url: str)

llama_cpp/llama_chat_format.py:3659

↓ 1 callersMethod_load_lora_adapters

(self, loras: List["Model.LoraAdapter"])

examples/server/server.py:11769

↓ 1 callersMethod_load_media_file

(self, kind: Literal["image", "audio", "video"], media_url: str)

examples/server/server.py:10634

↓ 1 callersMethod_load_media_url

(self, kind: Literal["image", "audio", "video"], media_url: str)

examples/server/server.py:10682

↓ 1 callersFunction_logit_bias_tokens_to_input_ids

( llama: llama_cpp.Llama, logit_bias: Dict[str, float], )

llama_cpp/server/app.py:225

↓ 1 callersMethod_media_identity_tokens

( self, kind: Literal["image", "audio", "video"], key: str, n_pos: int, )

examples/server/server.py:10836

↓ 1 callersMethod_messages_for_template

( messages: Sequence[ChatCompletionRequestMessage], media_marker: Optional[str], )

examples/server/server.py:3622

↓ 1 callersMethod_meta_value

(self, key: str)

examples/server/server.py:11909

↓ 1 callersMethod_metadata_compatible

(self, metadata: Optional[Dict[str, str]])

examples/server/server.py:12408

↓ 1 callersMethod_model_meta_key_by_index

(llama_model: Any, index: int)

examples/server/server.py:11830

↓ 1 callersMethod_needs_homogeneous_recurrent_draft_batching

(self)

examples/server/server.py:13656

↓ 1 callersMethod_new_stream_state

(self, plan: Dict[str, Any])

examples/server/server.py:5516

↓ 1 callersMethod_normalize_response_output_item_for_input

( item: Dict[str, Any], )

examples/server/server.py:8291

↓ 1 callersMethod_normalize_text_tool_payload

(payload: str)

examples/server/server.py:9339

↓ 1 callersMethod_oldest_entry

(self)

examples/server/server.py:12530

← previous next →301–400 of 1,336, ranked by callers