Merge branch 'upstream' into concedo_experimental

examples/main/main.cpp

@@ -32,8 +32,6 @@
 #pragma warning(disable: 4244 4267) // possible loss of data
 #endif
 
-static const char * DEFAULT_SYSTEM_MESSAGE = "You are a helpful assistant";
-
 static llama_context           ** g_ctx;
 static llama_model             ** g_model;
 static common_sampler          ** g_smpl;
@@ -220,6 +218,10 @@ int main(int argc, char ** argv) {
     // print chat template example in conversation mode
     if (params.conversation_mode) {
         if (params.enable_chat_template) {
+            if (!params.prompt.empty()) {
+                LOG_WRN("*** User-specified prompt in conversation mode will be ignored, did you mean to set --system-prompt (-sys) instead?\n");
+            }
+
             LOG_INF("%s: chat template example:\n%s\n", __func__, common_chat_format_example(chat_templates.get(), params.use_jinja).c_str());
         } else {
             LOG_INF("%s: in-suffix/prefix is specified, chat template will be disabled\n", __func__);
@@ -264,6 +266,7 @@ int main(int argc, char ** argv) {
 
     std::vector<llama_token> embd_inp;
 
+    bool waiting_for_first_input = params.conversation_mode && params.enable_chat_template && params.system_prompt.empty();
     auto chat_add_and_format = [&chat_msgs, &chat_templates](const std::string & role, const std::string & content) {
         common_chat_msg new_msg;
         new_msg.role = role;
@@ -275,11 +278,20 @@ int main(int argc, char ** argv) {
     };
 
     {
-        auto prompt = (params.conversation_mode && params.enable_chat_template)
-            // format the system prompt in conversation mode (fallback to default if empty)
-            ? chat_add_and_format("system", params.prompt.empty() ? DEFAULT_SYSTEM_MESSAGE : params.prompt)
+        std::string prompt;
+
+        if (params.conversation_mode && params.enable_chat_template) {
+            // format the system prompt in conversation mode (will use template default if empty)
+            prompt = params.system_prompt;
+
+            if (!prompt.empty()) {
+                prompt = chat_add_and_format("system", prompt);
+            }
+        } else {
             // otherwise use the prompt as is
-            : params.prompt;
+            prompt = params.prompt;
+        }
+
         if (params.interactive_first || !params.prompt.empty() || session_tokens.empty()) {
             LOG_DBG("tokenize the prompt\n");
             embd_inp = common_tokenize(ctx, prompt, true, true);
@@ -293,7 +305,7 @@ int main(int argc, char ** argv) {
     }
 
     // Should not run without any tokens
-    if (embd_inp.empty()) {
+    if (!params.conversation_mode && embd_inp.empty()) {
         if (add_bos) {
             embd_inp.push_back(llama_vocab_bos(vocab));
             LOG_WRN("embd_inp was considered empty and bos was added: %s\n", string_from(ctx, embd_inp).c_str());
@@ -477,8 +489,8 @@ int main(int argc, char ** argv) {
         LOG_INF(       " - Press Ctrl+C to interject at any time.\n");
 #endif
         LOG_INF(       "%s", control_message);
-        if (params.conversation_mode && params.enable_chat_template && params.prompt.empty()) {
-            LOG_INF(   " - Using default system message. To change it, set a different value via -p PROMPT or -f FILE argument.\n");
+        if (params.conversation_mode && params.enable_chat_template && params.system_prompt.empty()) {
+            LOG_INF(   " - Not using system message. To change it, set a different value via -sys PROMPT\n");
         }
         LOG_INF("\n");
 
@@ -774,7 +786,7 @@ int main(int argc, char ** argv) {
             }
 
             // deal with end of generation tokens in interactive mode
-            if (llama_vocab_is_eog(vocab, common_sampler_last(smpl))) {
+            if (!waiting_for_first_input && llama_vocab_is_eog(vocab, common_sampler_last(smpl))) {
                 LOG_DBG("found an EOG token\n");
 
                 if (params.interactive) {
@@ -794,12 +806,12 @@ int main(int argc, char ** argv) {
             }
 
             // if current token is not EOG, we add it to current assistant message
-            if (params.conversation_mode) {
+            if (params.conversation_mode && !waiting_for_first_input) {
                 const auto id = common_sampler_last(smpl);
                 assistant_ss << common_token_to_piece(ctx, id, false);
             }
 
-            if (n_past > 0 && is_interacting) {
+            if ((n_past > 0 || waiting_for_first_input) && is_interacting) {
                 LOG_DBG("waiting for user input\n");
 
                 if (params.conversation_mode) {
@@ -889,11 +901,12 @@ int main(int argc, char ** argv) {
                 input_echo = false; // do not echo this again
             }
 
-            if (n_past > 0) {
+            if (n_past > 0 || waiting_for_first_input) {
                 if (is_interacting) {
                     common_sampler_reset(smpl);
                 }
                 is_interacting = false;
+                waiting_for_first_input = false;
             }
         }
 

examples/server/tests/unit/test_chat_completion.py

@@ -144,6 +144,7 @@ def test_apply_chat_template():
 @pytest.mark.parametrize("response_format,n_predicted,re_content", [
     ({"type": "json_object", "schema": {"const": "42"}}, 6, "\"42\""),
     ({"type": "json_object", "schema": {"items": [{"type": "integer"}]}}, 10, "[ -3000 ]"),
+    ({"type": "json_schema", "json_schema": {"schema": {"const": "foooooo"}}}, 10, "\"foooooo\""),
     ({"type": "json_object"}, 10, "(\\{|John)+"),
     ({"type": "sound"}, 0, None),
     # invalid response format (expected to fail)

examples/server/tests/utils.py

@@ -26,7 +26,10 @@ from re import RegexFlag
 import wget
 
 
-DEFAULT_HTTP_TIMEOUT = 12 if "LLAMA_SANITIZE" not in os.environ else 30
+DEFAULT_HTTP_TIMEOUT = 12
+
+if "LLAMA_SANITIZE" in os.environ or "GITHUB_ACTION" in os.environ:
+    DEFAULT_HTTP_TIMEOUT = 30
 
 
 class ServerResponse:

examples/server/utils.hpp

@@ -590,8 +590,8 @@ static json oaicompat_completion_params_parse(
         if (response_type == "json_object") {
             json_schema = json_value(response_format, "schema", json::object());
         } else if (response_type == "json_schema") {
-            json json_schema = json_value(response_format, "json_schema", json::object());
-            json_schema = json_value(json_schema, "schema", json::object());
+            auto schema_wrapper = json_value(response_format, "json_schema", json::object());
+            json_schema = json_value(schema_wrapper, "schema", json::object());
         } else if (!response_type.empty() && response_type != "text") {
             throw std::runtime_error("response_format type must be one of \"text\" or \"json_object\", but got: " + response_type);
         }

examples/server/webui/src/components/ChatScreen.tsx

@@ -2,7 +2,7 @@ import { useEffect, useMemo, useRef, useState } from 'react';
 import { CallbackGeneratedChunk, useAppContext } from '../utils/app.context';
 import ChatMessage from './ChatMessage';
 import { CanvasType, Message, PendingMessage } from '../utils/types';
-import { classNames, throttle } from '../utils/misc';
+import { classNames, cleanCurrentUrl, throttle } from '../utils/misc';
 import CanvasPyInterpreter from './CanvasPyInterpreter';
 import StorageUtils from '../utils/storage';
 import { useVSCodeContext } from '../utils/llama-vscode';
@@ -18,6 +18,24 @@ export interface MessageDisplay {
   isPending?: boolean;
 }
 
+/**
+ * If the current URL contains "?m=...", prefill the message input with the value.
+ * If the current URL contains "?q=...", prefill and SEND the message.
+ */
+const prefilledMsg = {
+  content() {
+    const url = new URL(window.location.href);
+    return url.searchParams.get('m') ?? url.searchParams.get('q') ?? '';
+  },
+  shouldSend() {
+    const url = new URL(window.location.href);
+    return url.searchParams.has('q');
+  },
+  clear() {
+    cleanCurrentUrl(['m', 'q']);
+  },
+};
+
 function getListMessageDisplay(
   msgs: Readonly<Message[]>,
   leafNodeId: Message['id']
@@ -81,7 +99,7 @@ export default function ChatScreen() {
     canvasData,
     replaceMessageAndGenerate,
   } = useAppContext();
-  const [inputMsg, setInputMsg] = useState('');
+  const [inputMsg, setInputMsg] = useState(prefilledMsg.content());
   const inputRef = useRef<HTMLTextAreaElement>(null);
 
   const { extraContext, clearExtraContext } = useVSCodeContext(
@@ -172,6 +190,22 @@ export default function ChatScreen() {
 
   const hasCanvas = !!canvasData;
 
+  useEffect(() => {
+    if (prefilledMsg.shouldSend()) {
+      // send the prefilled message if needed
+      sendNewMessage();
+    } else {
+      // otherwise, focus on the input and move the cursor to the end
+      if (inputRef.current) {
+        inputRef.current.focus();
+        inputRef.current.selectionStart = inputRef.current.value.length;
+      }
+    }
+    prefilledMsg.clear();
+    // no need to keep track of sendNewMessage
+    // eslint-disable-next-line react-hooks/exhaustive-deps
+  }, [inputRef]);
+
   // due to some timing issues of StorageUtils.appendMsg(), we need to make sure the pendingMsg is not duplicated upon rendering (i.e. appears once in the saved conversation and once in the pendingMsg)
   const pendingMsgDisplay: MessageDisplay[] =
     pendingMsg && messages.at(-1)?.msg.id !== pendingMsg.id

examples/server/webui/src/components/SettingDialog.tsx

@@ -148,13 +148,13 @@ const SETTING_SECTIONS: SettingSection[] = [
     fields: [
       {
         type: SettingInputType.CHECKBOX,
-        label: 'Expand though process by default for generating message',
+        label: 'Expand thought process by default when generating messages',
         key: 'showThoughtInProgress',
       },
       {
         type: SettingInputType.CHECKBOX,
         label:
-          'Exclude thought process when sending request to API (Recommended for DeepSeek-R1)',
+          'Exclude thought process when sending requests to API (Recommended for DeepSeek-R1)',
         key: 'excludeThoughtOnReq',
       },
     ],
@@ -247,7 +247,7 @@ const SETTING_SECTIONS: SettingSection[] = [
               This feature uses{' '}
               <OpenInNewTab href="https://pyodide.org">pyodide</OpenInNewTab>,
               downloaded from CDN. To use this feature, ask the LLM to generate
-              python code inside a markdown code block. You will see a "Run"
+              Python code inside a Markdown code block. You will see a "Run"
               button on the code block, near the "Copy" button.
             </small>
           </>
@@ -274,7 +274,7 @@ export default function SettingDialog({
   );
 
   const resetConfig = () => {
-    if (window.confirm('Are you sure to reset all settings?')) {
+    if (window.confirm('Are you sure you want to reset all settings?')) {
       setLocalConfig(CONFIG_DEFAULT);
     }
   };
@@ -296,9 +296,9 @@ export default function SettingDialog({
           return;
         }
       } else if (mustBeNumeric) {
-        const trimedValue = value.toString().trim();
-        const numVal = Number(trimedValue);
-        if (isNaN(numVal) || !isNumeric(numVal) || trimedValue.length === 0) {
+        const trimmedValue = value.toString().trim();
+        const numVal = Number(trimmedValue);
+        if (isNaN(numVal) || !isNumeric(numVal) || trimmedValue.length === 0) {
           alert(`Value for ${key} must be numeric`);
           return;
         }

examples/server/webui/src/utils/misc.ts

@@ -118,3 +118,11 @@ export const throttle = <T extends unknown[]>(
     }, delay);
   };
 };
+
+export const cleanCurrentUrl = (removeQueryParams: string[]) => {
+  const url = new URL(window.location.href);
+  removeQueryParams.forEach((param) => {
+    url.searchParams.delete(param);
+  });
+  window.history.replaceState({}, '', url.toString());
+};

examples/tts/tts.cpp

@@ -3,6 +3,7 @@
 #include "sampling.h"
 #include "log.h"
 #include "llama.h"
+#include "json.hpp"
 
 #define _USE_MATH_DEFINES // For M_PI on MSVC
 
@@ -16,6 +17,13 @@
 #include <thread>
 #include <vector>
 
+using json = nlohmann::ordered_json;
+
+enum outetts_version {
+    OUTETTS_V0_2,
+    OUTETTS_V0_3,
+};
+
 //
 // Terminal utils
 //
@@ -371,7 +379,7 @@ static std::string replace_numbers_with_words(const std::string & input_text) {
 }
 
 // Based on: https://github.com/edwko/OuteTTS/blob/a613e79c489d8256dd657ea9168d78de75895d82/outetts/version/v1/prompt_processor.py#L39
-static std::string process_text(const std::string & text) {
+static std::string process_text(const std::string & text, const outetts_version tts_version = OUTETTS_V0_2) {
 
     // For now I skipped text romanization as I am unsure how to handle
     // uroman and MeCab implementations in C++
@@ -401,7 +409,8 @@ static std::string process_text(const std::string & text) {
         if (c == ' ') {
             prompt_clean += "<|text_sep|>";
     */
-    processed_text = std::regex_replace(processed_text, std::regex(R"(\s)"), "<|text_sep|>");
+    std::string separator = (tts_version == OUTETTS_V0_3) ? "<|space|>" : "<|text_sep|>";
+    processed_text = std::regex_replace(processed_text, std::regex(R"(\s)"), separator);
 
     return processed_text;
 }
@@ -425,8 +434,8 @@ static void prompt_init(llama_tokens & prompt, const llama_vocab * vocab) {
     prompt_add(prompt, vocab, "<|im_start|>\n", true, true);
 }
 
-static std::vector<llama_token> prepare_guide_tokens(const llama_vocab * vocab, const std::string & str) {
-    const std::string& delimiter = "<|text_sep|>";
+static std::vector<llama_token> prepare_guide_tokens(const llama_vocab * vocab, const std::string & str, const outetts_version tts_version = OUTETTS_V0_2) {
+    const std::string& delimiter = (tts_version == OUTETTS_V0_3 ? "<|space|>" : "<|text_sep|>");
 
     std::vector<llama_token> result;
     size_t start = 0;
@@ -452,6 +461,78 @@ static std::vector<llama_token> prepare_guide_tokens(const llama_vocab * vocab,
     return result;
 }
 
+static json speaker_from_file(const std::string & speaker_file) {
+    std::ifstream file(speaker_file);
+    if (!file) {
+        LOG_ERR("%s: Failed to open file '%s' for reading\n", __func__, speaker_file.c_str());
+        return json();
+    }
+
+    json speaker = json::parse(file);
+    return speaker;
+}
+
+static outetts_version get_tts_version(llama_model *model, json speaker = json::object()) {
+    if (speaker.contains("version")) {
+        std::string version = speaker["version"].get<std::string>();
+        if (version == "0.2") {
+            return OUTETTS_V0_2;
+        } else if (version == "0.3") {
+            return OUTETTS_V0_3;
+        } else {
+            LOG_ERR("%s: Unsupported speaker version '%s'\n", __func__, version.c_str());
+        }
+    }
+
+    // Also could get version from model itself
+    const char *chat_template = llama_model_chat_template(model, nullptr);
+    if (chat_template && std::string(chat_template) == "outetts-0.3") {
+        return OUTETTS_V0_3;
+    }
+
+    // Use 0.2 as the default version
+    return OUTETTS_V0_2;
+}
+
+static std::string audio_text_from_speaker(json speaker, const outetts_version tts_version = OUTETTS_V0_2) {
+    std::string audio_text = "<|text_start|>";
+
+    if (tts_version == OUTETTS_V0_2 || tts_version == OUTETTS_V0_3) {
+        std::string separator = (tts_version == OUTETTS_V0_3) ? "<|space|>" : "<|text_sep|>";
+        for (const auto &word : speaker["words"]) {
+            audio_text += word["word"].get<std::string>() + separator;
+        }
+    }
+
+    return audio_text;
+}
+
+static std::string audio_data_from_speaker(json speaker, const outetts_version tts_version = OUTETTS_V0_2) {
+    std::string audio_data = "<|audio_start|>\n";
+
+    if (tts_version == OUTETTS_V0_2 || tts_version == OUTETTS_V0_3) {
+        std::string code_start = (tts_version == OUTETTS_V0_3) ? "" : "<|code_start|>";
+        std::string code_end = (tts_version == OUTETTS_V0_3) ? "<|space|>" : "<|code_end|>";
+        for (const auto &word : speaker["words"]) {
+            std::string word_text = word["word"].get<std::string>();
+            double duration = word["duration"].get<double>();
+            std::vector<int> codes = word["codes"].get<std::vector<int>>();
+
+            // Create the audio output entry
+            std::ostringstream word_entry;
+            word_entry << word_text << "<|t_" << std::fixed << std::setprecision(2)
+                       << duration << "|>" + code_start;
+            for (const auto &Code : codes) {
+                word_entry << "<|" << Code << "|>";
+            }
+            word_entry << code_end << "\n";
+            audio_data += word_entry.str();
+        }
+    }
+
+    return audio_data;
+}
+
 int main(int argc, char ** argv) {
     common_params params;
 
@@ -523,34 +604,9 @@ int main(int argc, char ** argv) {
     std::vector<llama_token> codes;
     std::vector<llama_token> guide_tokens;
 
-    // process prompt and generate voice codes
-    {
-        LOG_INF("%s: constructing prompt ..\n", __func__);
-
-        std::vector<llama_token> prompt_inp;
-
-        prompt_init(prompt_inp, vocab);
-
-        prompt_add(prompt_inp, vocab, "<|text_start|>the<|text_sep|>overall<|text_sep|>package<|text_sep|>from<|text_sep|>just<|text_sep|>two<|text_sep|>people<|text_sep|>is<|text_sep|>pretty<|text_sep|>remarkable<|text_sep|>sure<|text_sep|>i<|text_sep|>have<|text_sep|>some<|text_sep|>critiques<|text_sep|>about<|text_sep|>some<|text_sep|>of<|text_sep|>the<|text_sep|>gameplay<|text_sep|>aspects<|text_sep|>but<|text_sep|>its<|text_sep|>still<|text_sep|>really<|text_sep|>enjoyable<|text_sep|>and<|text_sep|>it<|text_sep|>looks<|text_sep|>lovely<|text_sep|>", false, true);
-
-        // convert the input text into the necessary format expected by OuteTTS
-        {
-            std::string prompt_clean = process_text(params.prompt);
-            if (params.vocoder.use_guide_tokens) {
-                guide_tokens = prepare_guide_tokens(vocab, prompt_clean);
-            }
-
-            LOG_INF("%s: prompt: '%s'\n", __func__, prompt_clean.c_str());
-
-            prompt_add(prompt_inp, vocab, prompt_clean, false, true);
-        }
-
-        prompt_add(prompt_inp, vocab, "<|text_end|>\n", false, true);
-
-        // disabled to save time on tokenizing each time
-        // TODO: load voices from the json files
-#if 0
-        const std::string voice_data = R"(<|audio_start|>
+    // the default speaker profile is from: https://github.com/edwko/OuteTTS/blob/main/outetts/version/v1/default_speakers/en_male_1.json
+    std::string audio_text = "<|text_start|>the<|text_sep|>overall<|text_sep|>package<|text_sep|>from<|text_sep|>just<|text_sep|>two<|text_sep|>people<|text_sep|>is<|text_sep|>pretty<|text_sep|>remarkable<|text_sep|>sure<|text_sep|>i<|text_sep|>have<|text_sep|>some<|text_sep|>critiques<|text_sep|>about<|text_sep|>some<|text_sep|>of<|text_sep|>the<|text_sep|>gameplay<|text_sep|>aspects<|text_sep|>but<|text_sep|>its<|text_sep|>still<|text_sep|>really<|text_sep|>enjoyable<|text_sep|>and<|text_sep|>it<|text_sep|>looks<|text_sep|>lovely<|text_sep|>";
+    std::string audio_data = R"(<|audio_start|>
 the<|t_0.08|><|code_start|><|257|><|740|><|636|><|913|><|788|><|1703|><|code_end|>
 overall<|t_0.36|><|code_start|><|127|><|201|><|191|><|774|><|700|><|532|><|1056|><|557|><|798|><|298|><|1741|><|747|><|1662|><|1617|><|1702|><|1527|><|368|><|1588|><|1049|><|1008|><|1625|><|747|><|1576|><|728|><|1019|><|1696|><|1765|><|code_end|>
 package<|t_0.56|><|code_start|><|935|><|584|><|1319|><|627|><|1016|><|1491|><|1344|><|1117|><|1526|><|1040|><|239|><|1435|><|951|><|498|><|723|><|1180|><|535|><|789|><|1649|><|1637|><|78|><|465|><|1668|><|901|><|595|><|1675|><|117|><|1009|><|1667|><|320|><|840|><|79|><|507|><|1762|><|1508|><|1228|><|1768|><|802|><|1450|><|1457|><|232|><|639|><|code_end|>
@@ -582,117 +638,170 @@ it<|t_0.09|><|code_start|><|848|><|1366|><|395|><|1601|><|1513|><|593|><|1302|><
 looks<|t_0.27|><|code_start|><|1281|><|1266|><|1755|><|572|><|248|><|1751|><|1257|><|695|><|1380|><|457|><|659|><|585|><|1315|><|1105|><|1776|><|736|><|24|><|736|><|654|><|1027|><|code_end|>
 lovely<|t_0.56|><|code_start|><|634|><|596|><|1766|><|1556|><|1306|><|1285|><|1481|><|1721|><|1123|><|438|><|1246|><|1251|><|795|><|659|><|1381|><|1658|><|217|><|1772|><|562|><|952|><|107|><|1129|><|1112|><|467|><|550|><|1079|><|840|><|1615|><|1469|><|1380|><|168|><|917|><|836|><|1827|><|437|><|583|><|67|><|595|><|1087|><|1646|><|1493|><|1677|><|code_end|>)";
 
-        auto tmp = common_tokenize(vocab, voice_data, false, true);
-        printf("\n\n");
-        for (int i = 0; i < tmp.size(); ++i) {
-            printf("%d, ", tmp[i]);
+    // audio data for 0.3 version
+    outetts_version tts_version = get_tts_version(model_ttc);
+    if (tts_version == OUTETTS_V0_3) {
+        audio_text = std::regex_replace(audio_text, std::regex(R"(<\|text_sep\|>)"), "<|space|>");
+        audio_data = std::regex_replace(audio_data, std::regex(R"(<\|code_start\|>)"), "");
+        audio_data = std::regex_replace(audio_data, std::regex(R"(<\|code_end\|>)"), "<|space|>");
+    }
+
+    // load speaker if given
+    if (!params.vocoder.speaker_file.empty()) {
+        LOG_INF("%s: loading speaker ..\n", __func__);
+        json speaker = speaker_from_file(params.vocoder.speaker_file);
+        if (speaker.empty()) {
+            LOG_ERR("%s: Failed to load speaker file '%s'\n", __func__, params.vocoder.speaker_file.c_str());
+            return 1;
         }
-        printf("\n\n");
+        audio_text = audio_text_from_speaker(speaker, tts_version);
+        audio_data = audio_data_from_speaker(speaker, tts_version);
+    }
+
+    // process prompt and generate voice codes
+    {
+        LOG_INF("%s: constructing prompt ..\n", __func__);
+
+        std::vector<llama_token> prompt_inp;
+
+        prompt_init(prompt_inp, vocab);
+
+        prompt_add(prompt_inp, vocab, audio_text, false, true);
+
+        // convert the input text into the necessary format expected by OuteTTS
+        {
+            std::string prompt_clean = process_text(params.prompt, tts_version);
+            if (params.vocoder.use_guide_tokens) {
+                guide_tokens = prepare_guide_tokens(vocab, prompt_clean, tts_version);
+            }
+
+            LOG_INF("%s: prompt: '%s'\n", __func__, prompt_clean.c_str());
+
+            prompt_add(prompt_inp, vocab, prompt_clean, false, true);
+        }
+
+        prompt_add(prompt_inp, vocab, "<|text_end|>\n", false, true);
+
+        if (!params.vocoder.speaker_file.empty()) {
+            prompt_add(prompt_inp, vocab, audio_data, false, true);
+        } else {
+            // disabled to save time on tokenizing each time
+#if 1
+            const std::string voice_data = audio_data;
+
+            auto tmp = common_tokenize(vocab, voice_data, false, true);
+            printf("\n\n");
+            for (size_t i = 0; i < tmp.size(); ++i) {
+                printf("%d, ", tmp[i]);
+            }
+            printf("\n\n");
+            prompt_add(prompt_inp, tmp);
 #else
-        prompt_add(prompt_inp, llama_tokens {
-            151667, 198, 1782, 155780, 151669, 151929, 152412, 152308, 152585,
-            152460, 153375, 151670, 198, 74455, 155808, 151669, 151799,
-            151873, 151863, 152446, 152372, 152204, 152728, 152229, 152470,
-            151970, 153413, 152419, 153334, 153289, 153374, 153199, 152040,
-            153260, 152721, 152680, 153297, 152419, 153248, 152400, 152691,
-            153368, 153437, 151670, 198, 1722, 155828, 151669, 152607,
-            152256, 152991, 152299, 152688, 153163, 153016, 152789, 153198,
-            152712, 151911, 153107, 152623, 152170, 152395, 152852, 152207,
-            152461, 153321, 153309, 151750, 152137, 153340, 152573, 152267,
-            153347, 151789, 152681, 153339, 151992, 152512, 151751, 152179,
-            153434, 153180, 152900, 153440, 152474, 153122, 153129, 151904,
-            152311, 151670, 198, 1499, 155791, 151669, 152276, 152454,
-            153354, 152544, 153204, 153272, 152708, 153433, 152319, 153226,
-            153043, 152325, 153267, 152622, 151670, 198, 4250, 155797,
-            151669, 153454, 153342, 151989, 152458, 153420, 152303, 152271,
-            152827, 153036, 153196, 151708, 153263, 152561, 153207, 152213,
-            152112, 153204, 151722, 152542, 151670, 198, 19789, 155796,
-            151669, 153353, 153182, 152345, 152471, 152477, 153014, 152002,
-            152191, 151734, 152312, 152810, 152237, 153224, 153169, 153224,
-            152244, 153387, 153404, 151670, 198, 16069, 155811, 151669,
-            152265, 151946, 151808, 152412, 152363, 152305, 153156, 152733,
-            152810, 153157, 152016, 152100, 152069, 153234, 152317, 152589,
-            152707, 153121, 153341, 152159, 152114, 153156, 153001, 153504,
-            153376, 152272, 152433, 152325, 151941, 151670, 198, 285,
-            155788, 151669, 152238, 152255, 153427, 152318, 153009, 152381,
-            152474, 152680, 152157, 153255, 152324, 151682, 151670, 198,
-            32955, 155804, 151669, 153490, 153419, 152364, 152405, 152682,
-            152206, 152078, 153369, 152725, 153193, 153027, 152946, 152488,
-            153070, 151883, 152890, 152489, 153144, 153375, 152358, 151685,
-            152494, 152117, 152740, 151670, 198, 37448, 480, 155840, 151669,
-            151902, 152720, 153377, 152027, 152378, 152821, 153207, 153459,
-            153028, 153068, 152507, 153255, 152158, 152921, 151958, 152609,
-            152748, 152822, 152286, 151714, 152730, 152377, 152353, 152470,
-            152606, 152162, 152186, 153071, 152244, 153118, 153375, 153018,
-            152712, 153098, 152976, 152336, 151843, 153202, 152297, 151736,
-            153380, 153502, 152702, 152115, 153181, 152735, 153277, 153457,
-            152393, 153112, 152595, 151670, 198, 19098, 155808, 151669,
-            152464, 153452, 152595, 153312, 151937, 151933, 153197, 152239,
-            153163, 152922, 153402, 152034, 152591, 153438, 152215, 151673,
-            152005, 151785, 152642, 151924, 153278, 151805, 151974, 153482,
-            152718, 152862, 153347, 151670, 198, 72, 155780, 151669, 151795,
-            152111, 152746, 152377, 153471, 152309, 151670, 198, 19016,
-            155788, 151669, 153181, 152271, 152190, 152842, 152224, 152701,
-            152939, 152536, 152091, 151815, 152733, 151672, 151670, 198,
-            14689, 155788, 151669, 152291, 152072, 152942, 151734, 153042,
-            153504, 152589, 153333, 151839, 151941, 153038, 153180, 151670,
-            198, 36996, 8303, 155832, 151669, 152231, 152256, 152835,
-            152801, 152985, 153400, 152393, 152818, 152765, 152249, 152600,
-            151699, 152302, 152752, 153018, 153009, 151992, 153054, 152847,
-            153354, 153228, 152662, 153355, 152532, 153393, 151782, 152458,
-            152048, 152757, 152428, 153195, 151906, 153006, 153178, 153250,
-            152331, 152284, 152780, 153138, 153319, 151980, 153142, 152418,
-            152228, 152733, 151670, 198, 9096, 155801, 151669, 151698,
-            153321, 152217, 153039, 152935, 153400, 152122, 152531, 153106,
-            152169, 152892, 152957, 151851, 152427, 152826, 152451, 151851,
-            152901, 152885, 152594, 153446, 153080, 151670, 198, 14689,
-            155795, 151669, 152658, 151700, 153321, 152450, 152530, 153191,
-            151673, 151690, 151698, 152714, 152846, 152981, 153171, 153384,
-            153364, 153188, 153246, 151670, 198, 1055, 155779, 151669,
-            151869, 152388, 152711, 153334, 151736, 151670, 198, 1782,
-            155780, 151669, 153483, 153240, 152241, 152558, 152697, 153046,
-            151670, 198, 5804, 1363, 155820, 151669, 152941, 152764, 152605,
-            153034, 153434, 153372, 153347, 151887, 152453, 152758, 152133,
-            152510, 152694, 152431, 152321, 153088, 152676, 152223, 152581,
-            152459, 152015, 152502, 153063, 152712, 153294, 153451, 153032,
-            152903, 152859, 152989, 151748, 152669, 152661, 152650, 152409,
-            151861, 151670, 198, 300, 7973, 155828, 151669, 153095, 152469,
-            152988, 152894, 151819, 152391, 153019, 152058, 153062, 153230,
-            151826, 152112, 152306, 152264, 152769, 153390, 152384, 152435,
-            152790, 153393, 152983, 152540, 152252, 152034, 153107, 152540,
-            151919, 151893, 152558, 152817, 152946, 152956, 152129, 152715,
-            153131, 153490, 151734, 152271, 152707, 151734, 153321, 152450,
-            151670, 198, 8088, 155792, 151669, 152452, 153497, 153353,
-            152679, 152533, 152382, 152374, 152611, 153341, 153163, 152285,
-            153411, 152495, 153141, 152320, 151670, 198, 1199, 155781,
-            151669, 151764, 152360, 153295, 152634, 153342, 152199, 152271,
-            151670, 198, 43366, 155799, 151669, 152308, 151682, 152889,
-            152016, 152385, 152629, 152495, 151826, 153321, 152958, 152180,
-            151886, 153432, 152922, 152128, 153024, 153040, 152593, 152287,
-            151677, 151670, 198, 53660, 155808, 151669, 151727, 152092,
-            152680, 153331, 151699, 152316, 152938, 152289, 152433, 153384,
-            151781, 153137, 153259, 152175, 153213, 152291, 151869, 152691,
-            152489, 151941, 152049, 152034, 153053, 152179, 153160, 151676,
-            153367, 151670, 198, 268, 4123, 480, 155821, 151669, 152350,
-            152173, 152536, 151991, 151960, 153144, 153013, 152358, 152234,
-            153135, 152291, 153235, 152143, 152583, 152402, 153483, 152678,
-            152192, 152533, 152946, 151797, 153103, 152310, 152293, 151825,
-            152548, 153442, 152109, 152659, 153325, 152781, 152570, 152957,
-            151752, 152265, 153381, 152515, 151670, 198, 437, 155787,
-            151669, 152957, 152659, 151975, 152709, 152402, 152836, 152174,
-            151792, 153409, 153327, 152990, 151670, 198, 275, 155781,
-            151669, 152520, 153038, 152067, 153273, 153185, 152265, 152974,
-            151670, 198, 94273, 155799, 151669, 152953, 152938, 153427,
-            152244, 151920, 153423, 152929, 152367, 153052, 152129, 152331,
-            152257, 152987, 152777, 153448, 152408, 151696, 152408, 152326,
-            152699, 151670, 198, 385, 16239, 155828, 151669, 152306, 152268,
-            153438, 153228, 152978, 152957, 153153, 153393, 152795, 152110,
-            152918, 152923, 152467, 152331, 153053, 153330, 151889, 153444,
-            152234, 152624, 151779, 152801, 152784, 152139, 152222, 152751,
-            152512, 153287, 153141, 153052, 151840, 152589, 152508, 153499,
-            152109, 152255, 151739, 152267, 152759, 153318, 153165, 153349,
-            151670,});
+            prompt_add(prompt_inp, llama_tokens {
+                151667, 198, 1782, 155780, 151669, 151929, 152412, 152308, 152585,
+                152460, 153375, 151670, 198, 74455, 155808, 151669, 151799,
+                151873, 151863, 152446, 152372, 152204, 152728, 152229, 152470,
+                151970, 153413, 152419, 153334, 153289, 153374, 153199, 152040,
+                153260, 152721, 152680, 153297, 152419, 153248, 152400, 152691,
+                153368, 153437, 151670, 198, 1722, 155828, 151669, 152607,
+                152256, 152991, 152299, 152688, 153163, 153016, 152789, 153198,
+                152712, 151911, 153107, 152623, 152170, 152395, 152852, 152207,
+                152461, 153321, 153309, 151750, 152137, 153340, 152573, 152267,
+                153347, 151789, 152681, 153339, 151992, 152512, 151751, 152179,
+                153434, 153180, 152900, 153440, 152474, 153122, 153129, 151904,
+                152311, 151670, 198, 1499, 155791, 151669, 152276, 152454,
+                153354, 152544, 153204, 153272, 152708, 153433, 152319, 153226,
+                153043, 152325, 153267, 152622, 151670, 198, 4250, 155797,
+                151669, 153454, 153342, 151989, 152458, 153420, 152303, 152271,
+                152827, 153036, 153196, 151708, 153263, 152561, 153207, 152213,
+                152112, 153204, 151722, 152542, 151670, 198, 19789, 155796,
+                151669, 153353, 153182, 152345, 152471, 152477, 153014, 152002,
+                152191, 151734, 152312, 152810, 152237, 153224, 153169, 153224,
+                152244, 153387, 153404, 151670, 198, 16069, 155811, 151669,
+                152265, 151946, 151808, 152412, 152363, 152305, 153156, 152733,
+                152810, 153157, 152016, 152100, 152069, 153234, 152317, 152589,
+                152707, 153121, 153341, 152159, 152114, 153156, 153001, 153504,
+                153376, 152272, 152433, 152325, 151941, 151670, 198, 285,
+                155788, 151669, 152238, 152255, 153427, 152318, 153009, 152381,
+                152474, 152680, 152157, 153255, 152324, 151682, 151670, 198,
+                32955, 155804, 151669, 153490, 153419, 152364, 152405, 152682,
+                152206, 152078, 153369, 152725, 153193, 153027, 152946, 152488,
+                153070, 151883, 152890, 152489, 153144, 153375, 152358, 151685,
+                152494, 152117, 152740, 151670, 198, 37448, 480, 155840, 151669,
+                151902, 152720, 153377, 152027, 152378, 152821, 153207, 153459,
+                153028, 153068, 152507, 153255, 152158, 152921, 151958, 152609,
+                152748, 152822, 152286, 151714, 152730, 152377, 152353, 152470,
+                152606, 152162, 152186, 153071, 152244, 153118, 153375, 153018,
+                152712, 153098, 152976, 152336, 151843, 153202, 152297, 151736,
+                153380, 153502, 152702, 152115, 153181, 152735, 153277, 153457,
+                152393, 153112, 152595, 151670, 198, 19098, 155808, 151669,
+                152464, 153452, 152595, 153312, 151937, 151933, 153197, 152239,
+                153163, 152922, 153402, 152034, 152591, 153438, 152215, 151673,
+                152005, 151785, 152642, 151924, 153278, 151805, 151974, 153482,
+                152718, 152862, 153347, 151670, 198, 72, 155780, 151669, 151795,
+                152111, 152746, 152377, 153471, 152309, 151670, 198, 19016,
+                155788, 151669, 153181, 152271, 152190, 152842, 152224, 152701,
+                152939, 152536, 152091, 151815, 152733, 151672, 151670, 198,
+                14689, 155788, 151669, 152291, 152072, 152942, 151734, 153042,
+                153504, 152589, 153333, 151839, 151941, 153038, 153180, 151670,
+                198, 36996, 8303, 155832, 151669, 152231, 152256, 152835,
+                152801, 152985, 153400, 152393, 152818, 152765, 152249, 152600,
+                151699, 152302, 152752, 153018, 153009, 151992, 153054, 152847,
+                153354, 153228, 152662, 153355, 152532, 153393, 151782, 152458,
+                152048, 152757, 152428, 153195, 151906, 153006, 153178, 153250,
+                152331, 152284, 152780, 153138, 153319, 151980, 153142, 152418,
+                152228, 152733, 151670, 198, 9096, 155801, 151669, 151698,
+                153321, 152217, 153039, 152935, 153400, 152122, 152531, 153106,
+                152169, 152892, 152957, 151851, 152427, 152826, 152451, 151851,
+                152901, 152885, 152594, 153446, 153080, 151670, 198, 14689,
+                155795, 151669, 152658, 151700, 153321, 152450, 152530, 153191,
+                151673, 151690, 151698, 152714, 152846, 152981, 153171, 153384,
+                153364, 153188, 153246, 151670, 198, 1055, 155779, 151669,
+                151869, 152388, 152711, 153334, 151736, 151670, 198, 1782,
+                155780, 151669, 153483, 153240, 152241, 152558, 152697, 153046,
+                151670, 198, 5804, 1363, 155820, 151669, 152941, 152764, 152605,
+                153034, 153434, 153372, 153347, 151887, 152453, 152758, 152133,
+                152510, 152694, 152431, 152321, 153088, 152676, 152223, 152581,
+                152459, 152015, 152502, 153063, 152712, 153294, 153451, 153032,
+                152903, 152859, 152989, 151748, 152669, 152661, 152650, 152409,
+                151861, 151670, 198, 300, 7973, 155828, 151669, 153095, 152469,
+                152988, 152894, 151819, 152391, 153019, 152058, 153062, 153230,
+                151826, 152112, 152306, 152264, 152769, 153390, 152384, 152435,
+                152790, 153393, 152983, 152540, 152252, 152034, 153107, 152540,
+                151919, 151893, 152558, 152817, 152946, 152956, 152129, 152715,
+                153131, 153490, 151734, 152271, 152707, 151734, 153321, 152450,
+                151670, 198, 8088, 155792, 151669, 152452, 153497, 153353,
+                152679, 152533, 152382, 152374, 152611, 153341, 153163, 152285,
+                153411, 152495, 153141, 152320, 151670, 198, 1199, 155781,
+                151669, 151764, 152360, 153295, 152634, 153342, 152199, 152271,
+                151670, 198, 43366, 155799, 151669, 152308, 151682, 152889,
+                152016, 152385, 152629, 152495, 151826, 153321, 152958, 152180,
+                151886, 153432, 152922, 152128, 153024, 153040, 152593, 152287,
+                151677, 151670, 198, 53660, 155808, 151669, 151727, 152092,
+                152680, 153331, 151699, 152316, 152938, 152289, 152433, 153384,
+                151781, 153137, 153259, 152175, 153213, 152291, 151869, 152691,
+                152489, 151941, 152049, 152034, 153053, 152179, 153160, 151676,
+                153367, 151670, 198, 268, 4123, 480, 155821, 151669, 152350,
+                152173, 152536, 151991, 151960, 153144, 153013, 152358, 152234,
+                153135, 152291, 153235, 152143, 152583, 152402, 153483, 152678,
+                152192, 152533, 152946, 151797, 153103, 152310, 152293, 151825,
+                152548, 153442, 152109, 152659, 153325, 152781, 152570, 152957,
+                151752, 152265, 153381, 152515, 151670, 198, 437, 155787,
+                151669, 152957, 152659, 151975, 152709, 152402, 152836, 152174,
+                151792, 153409, 153327, 152990, 151670, 198, 275, 155781,
+                151669, 152520, 153038, 152067, 153273, 153185, 152265, 152974,
+                151670, 198, 94273, 155799, 151669, 152953, 152938, 153427,
+                152244, 151920, 153423, 152929, 152367, 153052, 152129, 152331,
+                152257, 152987, 152777, 153448, 152408, 151696, 152408, 152326,
+                152699, 151670, 198, 385, 16239, 155828, 151669, 152306, 152268,
+                153438, 153228, 152978, 152957, 153153, 153393, 152795, 152110,
+                152918, 152923, 152467, 152331, 153053, 153330, 151889, 153444,
+                152234, 152624, 151779, 152801, 152784, 152139, 152222, 152751,
+                152512, 153287, 153141, 153052, 151840, 152589, 152508, 153499,
+                152109, 152255, 151739, 152267, 152759, 153318, 153165, 153349,
+                151670,});
 #endif
+        }
 
         // print the prompt token-by-token