L

llm-cpp

Single-header C++17: token counts and dollar cost estimates for OpenAI and Anthropic models before you send, with budget checks. Part of llm-cpp.

Single-header C++17: A/B test prompts or models with Welch's t-test, Cohen's d and custom scorers. Part of llm-cpp.

Single-header C++17 libraries for LLM features: OpenAI and Anthropic streaming, retries, caching, cost estimates, RAG, structured JSON output and tool-calling agents. Copy one .hpp, no SDK, no package manager.

Single-header C++17: Whisper transcription and translation, and text-to-speech, via the OpenAI API. Part of llm-cpp.

Single-header C++17: send images (file or URL) plus a prompt to OpenAI or Anthropic vision models. Part of llm-cpp.

Single-header C++17: RAII spans with parent/child nesting, token and cost attributes, and OTLP-style JSON export. Part of llm-cpp.

Single-header C++17: Mustache-style prompt templates with loops, conditionals and token-budget truncation. Part of llm-cpp.

Single-header C++17: stream OpenAI and Anthropic chat responses token by token over SSE. Part of llm-cpp.

Single-header C++17: pick a model per prompt from a complexity score and a cost, latency, quality or budget strategy. Part of llm-cpp.

Single-header C++17: exponential backoff with jitter, provider failover and a circuit breaker for LLM calls. Part of llm-cpp.

Single-header C++17: rerank passages with offline BM25, LLM relevance scoring, or a hybrid of both. Part of llm-cpp.

Single-header C++17: end-to-end RAG that chunks, embeds, persists an index, retrieves top-k and answers. Part of llm-cpp.

Single-header C++17: worker pool with a priority queue and requests-per-minute and tokens-per-minute limits. Part of llm-cpp.

Single-header C++17: strip HTML and markdown, extract titles, links, headings and code blocks, and chunk text for LLMs. Part of llm-cpp.

Single-header C++17: fake LLM with scripted, pattern, random or echo responses, simulated latency and streaming, for tests. Part of llm-cpp.

Single-header C++17: structured JSONL log of every LLM call with latency, tokens and cost, plus query and summary. Part of llm-cpp.

Single-header C++17: small JSON parser and builder for LLM request bodies and model output. Part of llm-cpp.

Single-header C++17: detect and scrub PII (email, phone, SSN, card numbers, API keys) and score prompt-injection risk. Part of llm-cpp.

Single-header C++17: define a schema, validate model JSON against it, and re-prompt until the output conforms. Part of llm-cpp.

Single-header C++17: OpenAI fine-tuning lifecycle; write JSONL, upload, create, poll, cancel and list models. Part of llm-cpp.