AI Netcafe MCP Server

Translate a PDF from a URL while preserving the original layout — formulas, figures and two-column academic typesetting stay intact, unlike ordinary translators that flatten the document. Returns a job_id; poll check_job for the download links (translated-only and bilingual side-by-side). Typically 20-60 seconds for a few pages. Powered by PDFMathTranslate (36k stars) hosted at AI NetCafé.

Translate text between languages using a self-hosted LibreTranslate engine (fully offline, no big-tech API). For whole PDFs with layout preserved, use translate_pdf instead.

Full details of one hosted application: what it does, how to use it, measured benchmark scores, source repository, and the URL a human can open to run it.

List the open-source AI applications hosted and ready to run at AI NetCafé (ainetcafe.com). Each one normally requires local setup (Docker/Python + your own model API key); here they run pre-configured. Use this to find a tool for a task like translating a PDF with formulas intact, generating a PowerPoint file, polishing an academic paper, or running an autonomous research report. Do not call this first when the request already clearly matches compare_models, translate_pdf, deep_research, or make_slides; call that task tool directly.

Send a prompt to one specific large language model and get the answer plus its exact cost in USD. Useful when you want a second opinion from a different model, or a cheaper model for a bulk subtask.

Run one prompt across multiple LLMs in parallel and return every answer side by side with its real measured cost and latency. This answers "which model should I actually use for this kind of task?" with data instead of guesswork — useful before committing a long job to an expensive model.

Print-quality PDF from a URL or raw HTML via self-hosted Gotenberg (headless Chromium). Returns a hosted PDF download URL.

Start an autonomous web research task. The agent plans sub-questions, searches the web, reads the sources and writes a report with citations — this is real research, not a single model call, and takes 2-5 minutes. Returns a job_id immediately; poll check_job to get the report. Use this when you need sourced, current information rather than what a model already knows. Powered by gpt-researcher (29k stars) hosted at AI NetCafé.

Permanently delete one memory you own (id from recall). Use when a memory is wrong or sensitive.

Check text for grammar, spelling and style issues in 30+ languages (self-hosted LanguageTool). Returns each issue with a suggested replacement — apply them to produce corrected text.

Get the status or result of a job started by deep_research, translate_pdf, or make_slides. Poll every 15-30 seconds until status is "done" or "error". While work is pending, follow retry_after_seconds and next_action; when complete, prefer structured_result when present.

Turn a topic or an outline into a real downloadable .pptx file — not a link into someone's web editor. Returns a job_id; poll check_job for the download URL. Usually 1-3 minutes. Powered by Presenton (open source) hosted at AI NetCafé.

List every model callable through AI NetCafé with its input/output price per million tokens, so you can pick by cost as well as capability.

Fetch any public URL and return LLM-ready clean Markdown (rendered via Crawl4AI, handles JS pages). Use after web_search to read a source, or to ingest any page for analysis.

Check where a repository stands on AI NetCafé: live (with its public page and app URL), in the deployment pipeline, rejected (with the reason), or unknown. Use after submit_project.

Retrieve previously stored memories, optionally filtered by search query and/or project. Call at the start of work on a known project to restore context: why decisions were made, known fixes, preferences.

Persist a durable memory: an architecture decision, a stable user preference, a verified bug fix, or an important discovery. Anonymous callers get a small per-network memory pool; callers sending an AllRouter key (Authorization: Bearer sk-...) get a large pool shared across ALL their machines and agents — the same key on a laptop's Claude Code and a desktop's Codex recalls the same memories. Do not store secrets or raw logs.

Turn diagram-as-code into an image: Mermaid, PlantUML, Graphviz/DOT, C4, Excalidraw and 20+ more (self-hosted Kroki). Returns a hosted SVG/PNG URL you can embed directly in Markdown or HTML.

Submit a GitHub repository (a web app, typically AI-built) to the AI NetCafé hosting platform. The automated pipeline reviews it, containerizes it, deploys it on a dedicated subdomain with HTTPS and a pre-wired multi-LLM gateway, lists it in the store for humans, and exposes it to AI agents. Every use is metered and the author accrues a revenue share. Use this when a user says "deploy my project", "publish my app somewhere", or "I built something with AI, where can people use it?". Free to submit.

Fetch an audio file from a URL and transcribe it to text with open-source Whisper (100 languages, self-hosted). Good for voice memos, podcast clips and meeting recordings up to ~15 MB.

Search the live web through a self-hosted SearXNG meta-search (aggregates dozens of engines, no tracking). Returns titles, URLs and snippets. Use when you need current information or sources.