Method send_final_response

examples/server/server.cpp:1172–1217

    }

    void send_final_response(llama_client_slot &slot)
    {
        std::lock_guard<std::mutex> lock(mutex_results);
        task_result res;
        res.id = slot.task_id;
        res.error = false;
        res.stop = true;

        res.result_json = json
        {
            {"content",             !slot.params.stream ? slot.generated_text : ""},
            {"slot_id",             slot.id},
            {"stop",                true},
            {"model",               params.model_alias},
            {"tokens_predicted",    slot.n_decoded},
            {"tokens_evaluated",    slot.num_prompt_tokens},
            {"generation_settings", get_formated_generation(slot)},
            {"prompt",              slot.prompt},
            {"truncated",           slot.truncated},
            {"stopped_eos",         slot.stopped_eos},
            {"stopped_word",        slot.stopped_word},
            {"stopped_limit",       slot.stopped_limit},
            {"stopping_word",       slot.stopping_word},
            {"tokens_cached",       slot.n_past},
            {"timings",             slot.get_formated_timings()}
        };

        if (slot.sparams.n_probs > 0)
        {
            std::vector<completion_token_output> probs = {};
            if (!slot.params.stream && slot.stopped_word)
            {
                const std::vector<llama_token> stop_word_toks = llama_tokenize(ctx, slot.stopping_word, false);
                probs = std::vector<completion_token_output>(slot.generated_token_probs.begin(), slot.generated_token_probs.end() - stop_word_toks.size());
            }
            else
            {
                probs = std::vector<completion_token_output>(
                                    slot.generated_token_probs.begin(),
                                    slot.generated_token_probs.begin() + slot.sent_token_probs_index);
            }
            res.result_json["completion_probabilities"] = probs_vector_to_json(ctx, probs);
        }

        queue_results.push_back(res);
    }

    void send_embedding(llama_client_slot &slot)
    {