Modern C++23 LLM client built with modules —
importand chat, OpenAI-compatible, template-ready
| English - 简体中文 - 繁體中文 |
|---|
| mcpp build tool · package index · Documentation · Issues |
llmapi provides a typed Client<Provider> API for chat, streaming, embeddings, tool calls, and conversation persistence. The default config alias Config maps to OpenAI-style providers, so the common case does not need an explicit openai::OpenAI wrapper.
mcpp new myagent --template llmapi && cd myagent export OPENAI_API_KEY="sk-..." mcpp run
Templates ship with the library and version-track it automatically:
mcpp new --list-templates llmapi # list available templates mcpp new mybot --template llmapi:chat # interactive streaming chat CLI mcpp new mybot --template llmapi:anthropic # Anthropic provider mcpp new mybot --template llmapi:deepseek # OpenAI-compatible endpoint (DeepSeek)
Or add it to an existing mcpp project:
mcpp add llmapi
[dependencies.mcpplibs] llmapi = "0.2.8"
import mcpplibs.llmapi; import std; int main() { using namespace mcpplibs::llmapi; auto apiKey = std::getenv("OPENAI_API_KEY"); if (!apiKey) { std::println(stderr, "OPENAI_API_KEY not set"); return 1; } auto client = Client(Config{ .apiKey = apiKey, .model = "gpt-4o-mini", }); client.system("You are a concise assistant."); auto resp = client.chat("Explain why C++23 modules are useful in two sentences."); std::println("{}", resp.text()); return 0; }
import mcpplibs.llmapiwith C++23 modules- Strongly typed messages, tools, and response structs
- Sync, async, and streaming chat APIs
- Embeddings via the OpenAI provider
- Conversation save/load helpers
- OpenAI-compatible endpoint support through
openai::Config::baseUrl - Project templates:
mcpp new <name> --template llmapi[:<template>]
| Template | Description |
|---|---|
openai (default) |
Minimal OpenAI chat — one request, one answer |
chat |
Interactive streaming chat CLI (OpenAI) |
anthropic |
Anthropic (Claude) chat |
deepseek |
OpenAI-compatible endpoint via baseUrl (DeepSeek) |
openai::OpenAIfor OpenAI chat, streaming, embeddings, and OpenAI-compatible endpointsanthropic::Anthropicfor Anthropic chat and streamingConfigas a convenient alias foropenai::Config
Compatible endpoints can reuse the OpenAI provider:
auto client = Client(Config{ .apiKey = std::getenv("DEEPSEEK_API_KEY"), .baseUrl = std::string(URL::DeepSeek), .model = "deepseek-chat", });
git clone https://github.com/mcpplibs/llmapi.git && cd llmapi mcpp build
Apache-2.0 - see LICENSE