entropic::LlamaCppBackend Member List

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

activate()	entropic::InferenceBackend
allocate_temp_seq_id()	entropic::LlamaCppBackend	protected
apply_chat_template(const std::vector< Message > &messages, const GenerationParams &params) const	entropic::LlamaCppBackend	protected
capabilities() const	entropic::InferenceBackend
clear_prompt_cache() override	entropic::LlamaCppBackend	inlinevirtual
clear_state(int seq_id=-1)	entropic::InferenceBackend
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_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_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)	entropic::InferenceBackend	protectedvirtual
do_save_state(int seq_id, std::vector< uint8_t > &buffer) const	entropic::InferenceBackend	protectedvirtual
do_supports(BackendCapability cap) const override	entropic::LlamaCppBackend	protectedvirtual
do_unload() override	entropic::LlamaCppBackend	protectedvirtual
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_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_vision_	entropic::LlamaCppBackend	protected
info() const	entropic::InferenceBackend
init_mmproj_if_configured()	entropic::LlamaCppBackend	protected
is_active() const	entropic::InferenceBackend	inline
is_loaded() const	entropic::InferenceBackend	inline
is_recurrent() const	entropic::LlamaCppBackend	protected
is_recurrent_	entropic::LlamaCppBackend	protected
last_error_	entropic::InferenceBackend	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
prefill_and_cache_prefix(const std::vector< llama_token > &tokens, int prefix_tokens, const CacheKey &key)	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
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_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
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_hooks(const HookInterface &hooks)	entropic::InferenceBackend	inlineprotected
set_prompt_cache_config(const PromptCacheConfig &config)	entropic::LlamaCppBackend	inline
state() const	entropic::InferenceBackend	inline
step_token(llama_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
tokenize(const std::string &text, bool add_special) const	entropic::LlamaCppBackend	protected
tokenize_text(const std::string &text) const override	entropic::LlamaCppBackend	virtual
unload()	entropic::InferenceBackend
vocab_	entropic::LlamaCppBackend	protected
~InferenceBackend()=default (defined in entropic::InferenceBackend)	entropic::InferenceBackend	virtual
~LlamaCppBackend() override	entropic::LlamaCppBackend