tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-27 03:55:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

server: streaming of tool calls and thoughts when --jinja is on (#12379)

Browse Source 
* add common_json w/ support for truncated json healing

* add common_chat_msg_diff

* partial common_chat_parse

* refactor parser w/ optionals

* server: wire chat diffs in stream mode

* fix trigger of thinking models (must happen after thoughts are closed)

* fix functionary v3.2 raw python!

* rename: common_chat_syntax (now contains format)

* rm common_regex.at_start

* don't return empty <think></think>

* accommodate yet another deepseek r1 distill fantasy syntax (`<｜tool▁calls｜>`)

* fix QwQ 32B tool call parsing after thoughts (hermes2)

* better logs for grammar triggers

* consume spaces after parse_json_tool_calls

* fix required tool calls w/ thinking models that have pre-opened thinking tags

* fix thinking model's initial trigger + test qwq's template

* run most test_tool_call tests in stream + non-stream modes

* make functionary v3.2 parsing more strict (differentiate first match from others)

* send final diff from server, to close off raw python arguments

* support partial content streaming in Generic mode

* tool-call: allow content prelude before hermes2 tool calls (for Qwen2.5)

* Update function-calling.md

* Update tool_bench.py

* chat-parser: remove input from exception (llm output may contain PII)

---------

Co-authored-by: ochafik <ochafik@google.com>
Co-authored-by: Olivier Chafik <ochafik@users.noreply.github.com>
This commit is contained in:
Olivier Chafik
2025-05-25 01:48:08 +01:00
committed by GitHub
parent a2d02d5793
commit f5cd27b71d
23 changed files with 3245 additions and 1091 deletions
Download Patch File Download Diff File Expand all files Collapse all files

48
tools/server/utils.hpp
View File

@ -474,26 +474,6 @@ static std::string gen_tool_call_id() {
// other common utils
//
static bool ends_with(const std::string & str, const std::string & suffix) {
return str.size() >= suffix.size() && 0 == str.compare(str.size() - suffix.size(), suffix.size(), suffix);
}
static size_t find_partial_stop_string(const std::string &stop, const std::string &text) {
if (!text.empty() && !stop.empty()) {
const char text_last_char = text.back();
for (int64_t char_index = stop.size() - 1; char_index >= 0; char_index--) {
if (stop[char_index] == text_last_char) {
const std::string current_partial = stop.substr(0, char_index + 1);
if (ends_with(text, current_partial)) {
return text.size() - char_index - 1;
}
}
}
}
return std::string::npos;
}
// TODO: reuse llama_detokenize
template <class Iter>
static std::string tokens_to_str(llama_context * ctx, Iter begin, Iter end) {
@ -599,19 +579,16 @@ static json oaicompat_chat_params_parse(
json llama_params;
auto tools = json_value(body, "tools", json());
auto has_tools = tools.is_array() && !tools.empty();
auto stream = json_value(body, "stream", false);
auto tool_choice = json_value(body, "tool_choice", std::string("auto"));
if (tools.is_array() && !tools.empty()) {
if (stream) {
throw std::runtime_error("Cannot use tools with stream");
}
if (!opt.use_jinja) {
if (!opt.use_jinja) {
if (has_tools) {
throw std::runtime_error("tools param requires --jinja flag");
}
}
if (!opt.use_jinja) {
if (body.contains("tool_choice") && !body.at("tool_choice").is_null()) {
throw std::runtime_error("Unsupported param: tool_choice");
if (tool_choice != "auto") {
throw std::runtime_error("tool_choice param requires --jinja flag");
}
}
@ -749,14 +726,12 @@ static json oaicompat_chat_params_parse(
common_chat_templates_inputs inputs;
inputs.messages = common_chat_msgs_parse_oaicompat(messages);
inputs.tools = common_chat_tools_parse_oaicompat(tools);
inputs.tool_choice = common_chat_tool_choice_parse_oaicompat(json_value(body, "tool_choice", std::string("auto")));
inputs.tool_choice = common_chat_tool_choice_parse_oaicompat(tool_choice);
inputs.json_schema = json_schema.is_null() ? "" : json_schema.dump();
inputs.grammar = grammar;
inputs.add_generation_prompt = json_value(body, "add_generation_prompt", true);
inputs.use_jinja = opt.use_jinja;
inputs.parallel_tool_calls = json_value(body, "parallel_tool_calls", false);
inputs.extract_reasoning = opt.reasoning_format != COMMON_REASONING_FORMAT_NONE;
inputs.add_generation_prompt = json_value(body, "add_generation_prompt", true);
inputs.reasoning_format = opt.reasoning_format;
if (!inputs.tools.empty() && inputs.tool_choice != COMMON_CHAT_TOOL_CHOICE_NONE && body.contains("grammar")) {
throw std::runtime_error("Cannot use custom grammar constraints with tools.");
}
@ -774,7 +749,8 @@ static json oaicompat_chat_params_parse(
throw std::runtime_error("Cannot have 2 or more assistant messages at the end of the list.");
}
inputs.extract_reasoning = false;
/* TODO: test this properly */
inputs.reasoning_format = COMMON_REASONING_FORMAT_NONE;
inputs.add_generation_prompt = true;
}
@ -799,6 +775,7 @@ static json oaicompat_chat_params_parse(
}
llama_params["grammar_triggers"] = grammar_triggers;
llama_params["preserved_tokens"] = chat_params.preserved_tokens;
llama_params["thinking_forced_open"] = chat_params.thinking_forced_open;
for (const auto & stop : chat_params.additional_stops) {
llama_params["stop"].push_back(stop);
}
@ -812,6 +789,9 @@ static json oaicompat_chat_params_parse(
// Handle "logprobs" field
// TODO: The response format of this option is not yet OAI-compatible, but seems like no one really using it; We may need to fix it in the future
if (json_value(body, "logprobs", false)) {
if (has_tools && stream) {
throw std::runtime_error("logprobs is not supported with tools + stream");
}
llama_params["n_probs"] = json_value(body, "top_logprobs", 20);
} else if (body.contains("top_logprobs") && !body.at("top_logprobs").is_null()) {
throw std::runtime_error("top_logprobs requires logprobs to be set to true");