Single-turn question answering from a small local LLM (qwen3.5 2B). Best for general knowledge, definitions and simple reasoning. POST JSON {"input": string}, up to 3000 characters; returns {answer, model, elapsed_ms}. Answers are brief (2-3 sentences). Typically 5-15 s. Known limits: no web access, no memory between calls, and it can hallucinate or be wrong on recent events and obscure facts. Verify anything important.

Convert text into a 768-dimension embedding vector (nomic-embed-text) for semantic search, RAG retrieval, clustering, deduplication or similarity scoring. POST JSON {"input": string}, up to 3000 characters; returns {embedding: number[768], model, dimensions, elapsed_ms}. One text per call; English works best.

Summarize a passage of text into a short, concise summary with no preamble (qwen3.5 2B model). Good for condensing articles, emails, logs or documents before passing them to another agent. POST JSON {"input": string}, up to 3000 characters; returns {summary, model, elapsed_ms}. Typically 5-15 s. Small model: may miss nuance in technical or long text. Text only.