entropic::LlamaCppBackend Member List

This is the complete list of members for entropic::LlamaCppBackend, including all inherited members.

activate()	entropic::InferenceBackend
active_tools_json_	entropic::LlamaCppBackend	protected
allocate_temp_seq_id()	entropic::LlamaCppBackend	protected
allocate_temp_seq_id_for_test()	entropic::LlamaCppBackend	inline
apply_chat_template(const std::vector< Message > &messages, const GenerationParams &params) const	entropic::LlamaCppBackend	protected
apply_chat_template_lowlevel(const std::vector< Message > &messages) const	entropic::LlamaCppBackend	protected
build_batch_results(std::vector< BatchSeq > &seqs)	entropic::LlamaCppBackend	protected
build_mtp_head(const std::string &head_path)	entropic::LlamaCppBackend	protected
capabilities() const	entropic::InferenceBackend
clear_prompt_cache() override	entropic::LlamaCppBackend	inlinevirtual
clear_state(int seq_id=-1)	entropic::InferenceBackend
common_chat_parse_reliable() const	entropic::LlamaCppBackend
complete(const std::string &prompt, const GenerationParams &params)	entropic::InferenceBackend
compute_perplexity(const int32_t *tokens, int n_tokens)	entropic::InferenceBackend
compute_prefix_token_count(const std::vector< Message > &messages, const GenerationParams &params)	entropic::LlamaCppBackend	protected
config() const	entropic::InferenceBackend	inline
context_length() const	entropic::InferenceBackend	inline
count_tokens(const std::string &text) const	entropic::InferenceBackend
create_inference_context()	entropic::LlamaCppBackend	protected
create_sampler(const GenerationParams &params) const	entropic::LlamaCppBackend	protected
ctx_	entropic::LlamaCppBackend	protected
deactivate()	entropic::InferenceBackend
decode_loop(const std::vector< llama_token > &tokens, const GenerationParams &params, std::function< void(std::string_view)> on_token, std::atomic< bool > *cancel)	entropic::LlamaCppBackend	protected
decode_tokens_from(const std::vector< llama_token > &tokens, int start_offset)	entropic::LlamaCppBackend	protected
detokenize(llama_token token) const	entropic::LlamaCppBackend	protected
do_activate() override	entropic::LlamaCppBackend	protectedvirtual
do_backend_name() const override	entropic::LlamaCppBackend	protectedvirtual
do_clear_state(int seq_id) override	entropic::LlamaCppBackend	protectedvirtual
do_complete(const std::string &prompt, const GenerationParams &params) override	entropic::LlamaCppBackend	protectedvirtual
do_count_tokens(const std::string &text) const override	entropic::LlamaCppBackend	protectedvirtual
do_deactivate() override	entropic::LlamaCppBackend	protectedvirtual
do_evaluate_logprobs(const int32_t *tokens, int n_tokens) override	entropic::LlamaCppBackend	protectedvirtual
do_generate(const std::vector< Message > &messages, const GenerationParams &params) override	entropic::LlamaCppBackend	protectedvirtual
do_generate(const std::vector< Message > &messages, const GenerationParams &params, std::atomic< bool > &cancel) override	entropic::LlamaCppBackend	protectedvirtual
do_generate_batch(const std::vector< std::vector< Message > > &requests, const std::vector< GenerationParams > &params, std::atomic< bool > &cancel) override	entropic::LlamaCppBackend	protectedvirtual
do_generate_seq(int seq_id, const std::vector< Message > &messages, const GenerationParams &params)	entropic::InferenceBackend	protectedvirtual
do_generate_speculative(const std::vector< Message > &messages, const GenerationParams &params, std::function< void(std::string_view token)> on_token, std::atomic< bool > &cancel) override	entropic::LlamaCppBackend	protectedvirtual
do_generate_streaming(const std::vector< Message > &messages, const GenerationParams &params, std::function< void(std::string_view token)> on_token, std::atomic< bool > &cancel) override	entropic::LlamaCppBackend	protectedvirtual
do_generate_streaming_seq(int seq_id, const std::vector< Message > &messages, const GenerationParams &params, std::function< void(std::string_view token)> on_token, std::atomic< bool > &cancel)	entropic::InferenceBackend	protectedvirtual
do_generate_streaming_text_only(const std::vector< Message > &messages, const GenerationParams &params, std::function< void(std::string_view token)> on_token, std::atomic< bool > &cancel)	entropic::LlamaCppBackend	protected
do_generate_text_only(const std::vector< Message > &messages, const GenerationParams &params)	entropic::LlamaCppBackend	protected
do_generate_text_only(const std::vector< Message > &messages, const GenerationParams &params, std::atomic< bool > &cancel)	entropic::LlamaCppBackend	protected
do_info() const override	entropic::LlamaCppBackend	protectedvirtual
do_load(const ModelConfig &config) override	entropic::LlamaCppBackend	protectedvirtual
do_restore_state(int seq_id, const std::vector< uint8_t > &buffer) override	entropic::LlamaCppBackend	protectedvirtual
do_save_state(int seq_id, std::vector< uint8_t > &buffer) const override	entropic::LlamaCppBackend	protectedvirtual
do_supports(BackendCapability cap) const override	entropic::LlamaCppBackend	protectedvirtual
do_unload() override	entropic::LlamaCppBackend	protectedvirtual
effective_stop(const GenerationParams &params) const	entropic::LlamaCppBackend
evaluate_logprobs(const int32_t *tokens, int n_tokens)	entropic::InferenceBackend
extract_system_prompt(const std::vector< Message > &messages)	entropic::LlamaCppBackend	protectedstatic
extract_token_logprob(const float *logits, int32_t next_token, int n_vocab)	entropic::LlamaCppBackend	protectedstatic
fire_model_load_hook(const ModelConfig &config)	entropic::InferenceBackend	protected
free_seq_ids_	entropic::LlamaCppBackend	protected
generate(const std::vector< Message > &messages, const GenerationParams &params)	entropic::InferenceBackend
generate(const std::vector< Message > &messages, const GenerationParams &params, std::atomic< bool > &cancel)	entropic::InferenceBackend
generate_after_prefill(Sampler &sampler, const GenerationParams &params, std::function< void(std::string_view)> on_token, std::atomic< bool > *cancel)	entropic::LlamaCppBackend	protected
generate_batch(const std::vector< std::vector< Message > > &requests, const std::vector< GenerationParams > &params, std::atomic< bool > &cancel)	entropic::InferenceBackend
generate_mtp(const std::vector< Message > &messages, const GenerationParams &params, std::function< void(std::string_view token)> on_token, std::atomic< bool > &cancel, const std::string &head_path, int n_max)	entropic::LlamaCppBackend
generate_multimodal(const std::vector< Message > &messages, const GenerationParams &params, std::function< void(std::string_view token)> on_token, std::atomic< bool > *cancel)	entropic::LlamaCppBackend	protected
generate_seq(int seq_id, const std::vector< Message > &messages, const GenerationParams &params)	entropic::InferenceBackend
generate_speculative(const std::vector< Message > &messages, const GenerationParams &params, std::function< void(std::string_view token)> on_token, std::atomic< bool > &cancel)	entropic::InferenceBackend
generate_speculative_with_draft(const std::vector< Message > &messages, const GenerationParams &params, std::function< void(std::string_view token)> on_token, std::atomic< bool > &cancel, LlamaCppBackend &draft, int n_draft_max, const std::string &draft_path)	entropic::LlamaCppBackend
generate_streaming(const std::vector< Message > &messages, const GenerationParams &params, std::function< void(std::string_view token)> on_token, std::atomic< bool > &cancel)	entropic::InferenceBackend
generate_streaming_seq(int seq_id, const std::vector< Message > &messages, const GenerationParams &params, std::function< void(std::string_view token)> on_token, std::atomic< bool > &cancel)	entropic::InferenceBackend
has_common_chat_params() const	entropic::LlamaCppBackend	inline
has_vision_	entropic::LlamaCppBackend	protected
have_chat_params_	entropic::LlamaCppBackend	protected
info() const	entropic::InferenceBackend
init_mmproj_if_configured()	entropic::LlamaCppBackend	protected
inject_sampler_factory_for_test(std::unique_ptr< SamplerFactory > factory)	entropic::LlamaCppBackend
inject_tokenizer_for_test(std::unique_ptr< Tokenizer > tokenizer)	entropic::LlamaCppBackend
invalidate_resident_kv()	entropic::LlamaCppBackend	protected
is_active() const	entropic::InferenceBackend	inline
is_hybrid_	entropic::LlamaCppBackend	protected
is_loaded() const	entropic::InferenceBackend	inline
is_recurrent() const	entropic::LlamaCppBackend	protected
is_recurrent_	entropic::LlamaCppBackend	protected
kv_pos_max() const	entropic::LlamaCppBackend	inline
last_chat_format_	entropic::LlamaCppBackend	protected
last_error_	entropic::InferenceBackend	protected
last_gen_decode_calls() const	entropic::LlamaCppBackend	inline
last_gen_decode_calls_	entropic::LlamaCppBackend	protected
last_generation_prompt_	entropic::LlamaCppBackend	protected
last_input_tokens() const	entropic::LlamaCppBackend	inline
last_input_tokens_	entropic::LlamaCppBackend	protected
last_parser_	entropic::LlamaCppBackend	protected
last_prefill_ms() const	entropic::LlamaCppBackend	inline
last_prefill_ms_	entropic::LlamaCppBackend	protected
last_prefill_tokens() const	entropic::LlamaCppBackend	inline
last_prefill_tokens_	entropic::LlamaCppBackend	protected
llama_context_ptr()	entropic::LlamaCppBackend	inline
llama_model_ptr()	entropic::LlamaCppBackend	inline
load(const ModelConfig &config)	entropic::InferenceBackend
load_and_activate(const ModelConfig &config)	entropic::InferenceBackend
load_gpu_model()	entropic::LlamaCppBackend	protected
model_	entropic::LlamaCppBackend	protected
mtmd_ctx_	entropic::LlamaCppBackend	protected
mtmd_prefill(const std::string &prompt, const std::vector<::mtmd_bitmap * > &bitmaps, std::string &err_msg)	entropic::LlamaCppBackend	protected
mtp_active() const	entropic::LlamaCppBackend	inline
mtp_draft_ctx_	entropic::LlamaCppBackend	protected
mtp_draft_model_	entropic::LlamaCppBackend	protected
mtp_guard(const GenerationParams &params, const std::function< void(std::string_view)> &on_token, const std::string &head_path, int n_max)	entropic::LlamaCppBackend	protected
mtp_head_path_	entropic::LlamaCppBackend	protected
mtp_mutex_	entropic::LlamaCppBackend	protected
mtp_n_max_	entropic::LlamaCppBackend	protected
next_temp_seq_id_	entropic::LlamaCppBackend	protected
parse_chat_format_	entropic::LlamaCppBackend	protected
parse_generation_prompt_	entropic::LlamaCppBackend	protected
parse_params_valid_	entropic::LlamaCppBackend	protected
parse_parser_	entropic::LlamaCppBackend	protected
parse_response(const std::string &raw) const	entropic::LlamaCppBackend
prefill_and_cache_prefix(const std::vector< llama_token > &tokens, int prefix_tokens, const CacheKey &key)	entropic::LlamaCppBackend	protected
prefill_batch_suffixes(std::vector< BatchSeq > &seqs, const std::vector< std::vector< llama_token > > &toks, std::size_t shared)	entropic::LlamaCppBackend	protected
prefill_dispatch(const std::vector< llama_token > &tokens, const std::string &system_prompt, const std::vector< Message > &messages, const GenerationParams &params)	entropic::LlamaCppBackend	protected
prefill_shared_and_fanout(std::vector< BatchSeq > &seqs, const std::vector< llama_token > &seq0, std::size_t shared)	entropic::LlamaCppBackend	protected
prepare_batch_seqs(std::vector< BatchSeq > &seqs, const std::vector< GenerationParams > &params)	entropic::LlamaCppBackend	protected
prompt_cache_	entropic::LlamaCppBackend	protected
prompt_cache_config_	entropic::LlamaCppBackend	protected
release_temp_seq_id(llama_seq_id seq_id)	entropic::LlamaCppBackend	protected
release_temp_seq_id_for_test(llama_seq_id id)	entropic::LlamaCppBackend	inline
release_temp_seqs(std::vector< BatchSeq > &seqs)	entropic::LlamaCppBackend	protected
reload_model_cpu_only()	entropic::LlamaCppBackend	protected
render_prompt(const std::vector< Message > &messages, const GenerationParams &params)	entropic::LlamaCppBackend	protected
render_with_tools(const std::vector< Message > &messages, const GenerationParams &params)	entropic::LlamaCppBackend
resident_tokens_	entropic::LlamaCppBackend	protected
restore_cached_prefix(const CacheEntry *cached, const std::vector< llama_token > &tokens)	entropic::LlamaCppBackend	protected
restore_state(int seq_id, const std::vector< uint8_t > &buffer)	entropic::InferenceBackend
run_batch_gen_loop(std::vector< BatchSeq > &seqs, int max_steps, std::atomic< bool > &cancel)	entropic::LlamaCppBackend	protected
run_batched_decode(const std::vector< std::vector< llama_token > > &toks, const std::vector< GenerationParams > &params, std::size_t shared, std::atomic< bool > &cancel)	entropic::LlamaCppBackend	protected
run_prefill(const std::vector< llama_token > &tokens)	entropic::LlamaCppBackend	protected
run_prefill_cached(const std::vector< llama_token > &tokens, const std::string &system_prompt, const std::vector< Message > &messages, const GenerationParams &params)	entropic::LlamaCppBackend	protected
run_sampling_loop(const GenerationParams &params, std::function< void(std::string_view token)> on_token, std::atomic< bool > *cancel, const std::chrono::steady_clock::time_point &t0)	entropic::LlamaCppBackend	protected
sample_batch_active(std::vector< BatchSeq > &seqs)	entropic::LlamaCppBackend	protected
sampler_factory_	entropic::LlamaCppBackend	protected
sampler_factory_for_test() const	entropic::LlamaCppBackend	inline
save_prefix_to_cache(const CacheKey &key, int prefix_tokens)	entropic::LlamaCppBackend	protected
save_state(int seq_id, std::vector< uint8_t > &buffer) const	entropic::InferenceBackend
seq_id_mutex_	entropic::LlamaCppBackend	protected
set_active_tools(const std::string &tools_json)	entropic::LlamaCppBackend
set_hooks(const HookInterface &hooks)	entropic::InferenceBackend	inlineprotected
set_prompt_cache_config(const PromptCacheConfig &config)	entropic::LlamaCppBackend	inline
setup_mtp_draft(const std::string &head_path, int n_max)	entropic::LlamaCppBackend	protected
state() const	entropic::InferenceBackend	inline
state_	entropic::InferenceBackend	protected
step_token(Sampler &sampler, std::string &generated, std::function< void(std::string_view)> &on_token, const std::vector< std::string > &stop)	entropic::LlamaCppBackend	protected
supports(BackendCapability cap) const	entropic::InferenceBackend
teardown_mtp_draft()	entropic::LlamaCppBackend	protected
tokenize(const std::string &text, bool add_special) const	entropic::LlamaCppBackend	protected
tokenize_text(const std::string &text) const override	entropic::LlamaCppBackend	virtual
tokenizer_	entropic::LlamaCppBackend	protected
tool_call_close_marker() const override	entropic::LlamaCppBackend	virtual
try_warm_reuse(const std::vector< llama_token > &tokens)	entropic::LlamaCppBackend	protected
unload()	entropic::InferenceBackend
vocab_	entropic::LlamaCppBackend	protected
~InferenceBackend()=default (defined in entropic::InferenceBackend)	entropic::InferenceBackend	virtual
~LlamaCppBackend() override	entropic::LlamaCppBackend