mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-07-23 11:10:55 -05:00
chat : implement minicpm5 parser (#24889)
* Add minicpm5 tool call parser * Refactor MiniCPM5 PEG parser per review feedback * Fix jinja min/max API to match Jinja2 * modify by review * MiniCPM5: use autoparser for XML tool calls and fix grammar preserved-token triggers * MiniCPM5: fix streaming tool-arg placeholder and remove alt XML markers * skip min/max attribute tests in -py mode * test-jinja: use real expected output for min/max attribute tests * MiniCPM5: revert shared mapper and history fallbacks per review Drop streaming tool-arg placeholder workarounds from the generic PEG mapper and restore strict tool-call argument JSON parsing so MiniCPM5 support stays limited to autoparser/diff-analyzer changes. * chat : refactor minicpm5 back to dedicated parser * cont : simplify grammar * cont : refactor * cont : fixes * cont : rename template to openbmb-MiniCPM5-1B.jinja * cont : add message delimiters * cont : fix tests --------- Co-authored-by: zhangtao <zhangtao2@modelbest.cn> Co-authored-by: 张涛 <>
This commit is contained in:
151
common/chat.cpp
151
common/chat.cpp
@@ -2376,6 +2376,149 @@ static void func_args_not_string(json & messages) {
|
||||
|
||||
}
|
||||
|
||||
// MiniCPM5 format:
|
||||
// - Reasoning: <think>{reasoning}</think> (optional)
|
||||
// - Tool calls: <function name="foo"><param name="bar">value</param></function>
|
||||
static common_chat_params common_chat_params_init_minicpm5(const common_chat_template & tmpl,
|
||||
const autoparser::generation_params & inputs) {
|
||||
common_chat_params data;
|
||||
|
||||
data.prompt = common_chat_template_direct_apply_impl(tmpl, inputs);
|
||||
data.generation_prompt = common_chat_template_generation_prompt_impl(tmpl, inputs);
|
||||
data.format = COMMON_CHAT_FORMAT_PEG_NATIVE;
|
||||
data.supports_thinking = true;
|
||||
data.preserved_tokens = {
|
||||
"<function",
|
||||
"<param",
|
||||
"</function>",
|
||||
"</param>",
|
||||
"<think>",
|
||||
"</think>",
|
||||
};
|
||||
|
||||
data.thinking_start_tag = "<think>";
|
||||
data.thinking_end_tag = "</think>";
|
||||
|
||||
data.message_delimiters = {
|
||||
{ COMMON_CHAT_ROLE_ASSISTANT, "<|im_start|>assistant" },
|
||||
{ COMMON_CHAT_ROLE_TOOL, "<|im_start|>user\n<tool_response>" },
|
||||
{ COMMON_CHAT_ROLE_USER, "<|im_start|>user" },
|
||||
{ COMMON_CHAT_ROLE_SYSTEM, "<|im_start|>system" },
|
||||
};
|
||||
|
||||
auto has_tools = inputs.tools.is_array() && !inputs.tools.empty();
|
||||
auto has_response_format = inputs.json_schema.is_object() && !inputs.json_schema.empty();
|
||||
auto extract_reasoning = inputs.reasoning_format != COMMON_REASONING_FORMAT_NONE;
|
||||
auto include_grammar = has_response_format || (has_tools && inputs.tool_choice != COMMON_CHAT_TOOL_CHOICE_NONE);
|
||||
|
||||
if (inputs.has_continuation()) {
|
||||
const auto & msg = inputs.continue_msg;
|
||||
|
||||
data.generation_prompt = "<|im_start|>assistant\n<think>\n" + msg.reasoning_content;
|
||||
if (inputs.continue_final_message == COMMON_CHAT_CONTINUATION_CONTENT) {
|
||||
data.generation_prompt += "\n</think>\n\n" + msg.render_content();
|
||||
}
|
||||
|
||||
data.prompt += data.generation_prompt;
|
||||
}
|
||||
|
||||
auto parser = build_chat_peg_parser([&](common_chat_peg_builder & p) {
|
||||
auto generation_prompt = p.literal("<|im_start|>assistant\n");
|
||||
|
||||
auto reasoning = p.eps();
|
||||
if (extract_reasoning) {
|
||||
reasoning = ("<think>" << p.reasoning(p.until("</think>")) << "</think>") + p.space();
|
||||
}
|
||||
|
||||
// Response format parser
|
||||
if (has_response_format) {
|
||||
return generation_prompt + reasoning + p.content(p.schema(p.json(), "response-format", inputs.json_schema));
|
||||
}
|
||||
|
||||
if (has_tools && inputs.tool_choice != COMMON_CHAT_TOOL_CHOICE_NONE) {
|
||||
// CDATA lets a value carry characters that would otherwise close the tag (e.g.
|
||||
// </param>); capture the inner text only, excluding the CDATA markers.
|
||||
auto string_value = p.choice({
|
||||
p.literal("<![CDATA[") + p.ac(p.tool_arg_string_value(p.until("]]>")) + p.literal("]]>"), "]]>") + p.tool_arg_close(p.literal("</param>")),
|
||||
p.negate(p.literal("< {
|
||||
const auto & function = tool.at("function");
|
||||
const std::string name = function.at("name");
|
||||
auto params = function.contains("parameters") ? function.at("parameters") : json::object();
|
||||
|
||||
auto args = p.eps();
|
||||
if (params.contains("properties") && params.at("properties").is_object() && !params.at("properties").empty()) {
|
||||
auto schema_info = common_schema_info();
|
||||
schema_info.resolve_refs(params);
|
||||
|
||||
auto arg_choice = p.choice();
|
||||
for (const auto & [prop_name, prop_schema] : params.at("properties").items()) {
|
||||
auto value_parser = p.eps();
|
||||
if (schema_info.resolves_to_string(prop_schema)) {
|
||||
value_parser = string_value;
|
||||
} else {
|
||||
value_parser = p.tool_arg_json_value(
|
||||
p.schema(p.json(), "tool-" + name + "-arg-" + prop_name + "-schema", prop_schema, false)
|
||||
) + p.tool_arg_close(p.literal("</param>"));
|
||||
}
|
||||
|
||||
auto arg_rule = p.tool_arg(
|
||||
p.tool_arg_open(p.literal("<param name=\"") + p.tool_arg_name(p.literal(prop_name)) + p.literal("\">")) +
|
||||
value_parser
|
||||
);
|
||||
|
||||
arg_choice |= arg_rule;
|
||||
}
|
||||
args = p.zero_or_more(arg_choice + p.space());
|
||||
}
|
||||
|
||||
auto tool_parser = p.tool(
|
||||
p.tool_open(p.literal("<function name=\"") + p.tool_name(p.literal(name)) + p.literal("\">"))
|
||||
<< p.tool_args(args)
|
||||
<< p.tool_close(p.literal("</function>")));
|
||||
|
||||
tool_choice |= p.rule("tool-" + name, tool_parser);
|
||||
});
|
||||
|
||||
auto max_calls = inputs.parallel_tool_calls ? -1 : 1;
|
||||
auto tool_calls = p.trigger_rule("tool-call", p.repeat(tool_choice + p.space(), 1, max_calls));
|
||||
|
||||
auto content = p.content(p.until("<function"));
|
||||
|
||||
return generation_prompt + reasoning + content + tool_calls + p.end();
|
||||
}
|
||||
|
||||
return generation_prompt + reasoning + p.content(p.rest()) + p.end();
|
||||
});
|
||||
|
||||
data.parser = parser.save();
|
||||
|
||||
if (include_grammar) {
|
||||
data.grammar_lazy = !(has_response_format || (has_tools && inputs.tool_choice == COMMON_CHAT_TOOL_CHOICE_REQUIRED));
|
||||
data.grammar = build_grammar([&](const common_grammar_builder & builder) {
|
||||
foreach_function(inputs.tools, [&](const json & tool) {
|
||||
const auto & function = tool.at("function");
|
||||
auto schema = function.contains("parameters") ? function.at("parameters") : json::object();
|
||||
builder.resolve_refs(schema);
|
||||
});
|
||||
if (has_response_format) {
|
||||
auto schema = inputs.json_schema;
|
||||
builder.resolve_refs(schema);
|
||||
}
|
||||
parser.build_grammar(builder, data.grammar_lazy);
|
||||
});
|
||||
|
||||
data.grammar_triggers = {
|
||||
{ COMMON_GRAMMAR_TRIGGER_TYPE_WORD, "<function" },
|
||||
};
|
||||
}
|
||||
|
||||
return data;
|
||||
}
|
||||
|
||||
static json common_chat_extra_context() {
|
||||
json ctx = json::object();
|
||||
std::chrono::system_clock::time_point now = std::chrono::system_clock::now();
|
||||
@@ -2468,6 +2611,14 @@ std::optional<common_chat_params> common_chat_try_specialized_template(
|
||||
return common_chat_params_init_gemma4(tmpl, params);
|
||||
}
|
||||
|
||||
// MiniCPM5 - XML tool calls with <function name="..."><param name="...">...</param></function>
|
||||
if (src.find("Tool usage guidelines:") != std::string::npos &&
|
||||
src.find("<function name=\"") != std::string::npos &&
|
||||
src.find("<param name=\"") != std::string::npos) {
|
||||
LOG_DBG("Using specialized template: MiniCPM5\n");
|
||||
return common_chat_params_init_minicpm5(tmpl, params);
|
||||
}
|
||||
|
||||
return std::nullopt;
|
||||
}
|
||||
|
||||
|
||||
@@ -1108,6 +1108,50 @@ const func_builtins & value_array_t::get_builtins() const {
|
||||
std::reverse(arr.begin(), arr.end());
|
||||
return is_val<value_tuple>(val) ? mk_val<value_tuple>(std::move(arr)) : mk_val<value_array>(std::move(arr));
|
||||
}},
|
||||
{"min", [](const func_args & args) -> value {
|
||||
args.ensure_count(1, 4);
|
||||
args.ensure_vals<value_array>();
|
||||
value val_case = args.get_kwarg_or_pos("case_sensitive", 1);
|
||||
value attribute = args.get_kwarg_or_pos("attribute", 2);
|
||||
if (!attribute->is_undefined()) {
|
||||
throw not_implemented_exception("min: attribute not implemented");
|
||||
}
|
||||
// FIXME: min is currently always case sensitive
|
||||
(void) val_case;
|
||||
const auto & arr = args.get_pos(0)->as_array();
|
||||
if (arr.empty()) {
|
||||
return mk_val<value_undefined>();
|
||||
}
|
||||
value result = arr[0];
|
||||
for (size_t i = 1; i < arr.size(); ++i) {
|
||||
if (value_compare(arr[i], result, value_compare_op::lt)) {
|
||||
result = arr[i];
|
||||
}
|
||||
}
|
||||
return result;
|
||||
}},
|
||||
{"max", [](const func_args & args) -> value {
|
||||
args.ensure_count(1, 4);
|
||||
args.ensure_vals<value_array>();
|
||||
value val_case = args.get_kwarg_or_pos("case_sensitive", 1);
|
||||
value attribute = args.get_kwarg_or_pos("attribute", 2);
|
||||
if (!attribute->is_undefined()) {
|
||||
throw not_implemented_exception("max: attribute not implemented");
|
||||
}
|
||||
// FIXME: max is currently always case sensitive
|
||||
(void) val_case;
|
||||
const auto & arr = args.get_pos(0)->as_array();
|
||||
if (arr.empty()) {
|
||||
return mk_val<value_undefined>();
|
||||
}
|
||||
value result = arr[0];
|
||||
for (size_t i = 1; i < arr.size(); ++i) {
|
||||
if (value_compare(arr[i], result, value_compare_op::gt)) {
|
||||
result = arr[i];
|
||||
}
|
||||
}
|
||||
return result;
|
||||
}},
|
||||
{"unique", array_unique_not_implemented},
|
||||
};
|
||||
return builtins;
|
||||
|
||||
Reference in New Issue
Block a user