llama.cpp/common/chat.hpp
Olivier Chafik c7f460ab88
server: fix tool-call of DeepSeek R1 Qwen, return reasoning_content (Command 7RB & DeepSeek R1) unless --reasoning-format none (#11607)
* extract & return thoughts in reasoning_content field (unless --reasoning-format) for DeepSeek R1 & Command R7B

* tool-calls: add deepseek r1 template (models/templates/llama-cpp-deepseek-r1.jinja) + hackommodate broken official template

* tool-calls: accommodate variety of wrong tool call opening tags both R1 Qwen 32B and 7B distills like to spit out

* server/oai: ensure content is null when there are tool calls, and reasoning_content appears before content for readability

* tool-calls: add DeepSeek R1 Qwen distills to server/README.md & server tests

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2025-02-13 10:05:16 +00:00

56 lines
1.9 KiB
C++

// Chat support (incl. tool call grammar constraining & output parsing) w/ generic & custom template handlers.
#pragma once
#include "common.h"
#include <json.hpp>
#include <optional>
#include <string>
#include <vector>
using json = nlohmann::ordered_json;
struct common_chat_inputs {
json messages;
json tools;
json tool_choice;
json json_schema;
bool parallel_tool_calls;
bool stream;
std::string grammar;
bool add_generation_prompt = true;
bool extract_reasoning = true;
};
enum common_chat_format {
COMMON_CHAT_FORMAT_CONTENT_ONLY,
COMMON_CHAT_FORMAT_GENERIC,
COMMON_CHAT_FORMAT_MISTRAL_NEMO,
COMMON_CHAT_FORMAT_LLAMA_3_X,
COMMON_CHAT_FORMAT_LLAMA_3_X_WITH_BUILTIN_TOOLS,
COMMON_CHAT_FORMAT_DEEPSEEK_R1,
COMMON_CHAT_FORMAT_DEEPSEEK_R1_EXTRACT_REASONING,
COMMON_CHAT_FORMAT_FIREFUNCTION_V2,
COMMON_CHAT_FORMAT_FUNCTIONARY_V3_2,
COMMON_CHAT_FORMAT_FUNCTIONARY_V3_1_LLAMA_3_1,
COMMON_CHAT_FORMAT_HERMES_2_PRO,
COMMON_CHAT_FORMAT_COMMAND_R7B,
COMMON_CHAT_FORMAT_COMMAND_R7B_EXTRACT_REASONING,
COMMON_CHAT_FORMAT_COUNT, // Not a format, just the # formats
};
struct common_chat_params {
common_chat_format format = COMMON_CHAT_FORMAT_CONTENT_ONLY;
json prompt;
std::string grammar;
bool grammar_lazy = false;
std::vector<common_grammar_trigger> grammar_triggers;
std::vector<std::string> preserved_tokens;
std::vector<std::string> additional_stops;
};
struct common_chat_params common_chat_params_init(const common_chat_template & tmpl, const struct common_chat_inputs & params);
std::string common_chat_format_name(common_chat_format format);
common_chat_msg common_chat_parse( const std::string & input, common_chat_format format);