⏱️ AI PRIVACY SPEED TEST: Maple AI vs MacBook Pro. We give Maple the larger GPT-OSS 120B model and a MBP 36GB the smaller GPT-OSS 20B from @OpenAI. The test: "Build me a business plan for a lemonade stand". Can Maple keep up while running a model with 6x the parameters? Can

RT @moneyball: Goose 🤝 Maple.

RT @MartyBent: If your AI isn’t using confidential compute all of your chats are completely exposed to the company hosting the models. Use….

RT @MaxAWebster: I'm insanely bullish on confidential compute and @TryMapleAI / @OpenSecretCloud . AI will provide a tremendous amount of v….

Read the announcement blog post

Start building with private AI today: download the app or pull the Docker image, upgrade to a plan, add a few dollars of credits, point your client to  http://localhost:8080/v1, and secure all your apps.

Need more detail? Check the technical write‑up, full API reference on GitHub, or join the Discord community for real‑time help.

Compatibility: Any library that lets you set a base URL works—LangChain, LlamaIndex, Amp, Open Interpreter, Goose, Jan, and virtually every OpenAI‑compatible SDK. Here it powers Goose by @blocks

Docker image (production‑ready). - docker pull - Run with your MAPLE_API_KEY and MAPLE_BACKEND_URL.- You now have a secure OpenAI‑compatible endpoint at  http://localhost:8080/v1.

Getting started is simple:. Desktop app (fastest for local dev). - Download from - Sign up for a Pro/Team/Max plan (starts at $20/mo).- Purchase $10+ of credits.- Click “Start Proxy” → API key & localhost endpoint are ready.

👩‍💻 A coding‑assistant plug‑in for any IDE points to  http://localhost:8080/v1  and suggests code, refactors, and explains errors without exposing proprietary code. Vibe coding with encryption.

📚 A startup’s internal knowledge‑base search runs through the proxy, so confidential architecture details never leave the enclave.

Real‑world use cases: .🗓️ A calorie‑counting app replaces public OpenAI calls with Maple Proxy, delivering personalized meal plans while keeping dietary data private.

Models (pay‑as‑you‑go):.- llama3‑3‑70b.- gpt‑oss‑120b.- deepseek‑r1‑0528 .- mistral‑small‑3‑1‑24b.- qwen2‑5‑72b .- qwen3‑coder‑480bt.- gemma‑3‑27b‑it‑fp8‑dynamic.

How it works: .🔒 Inference runs inside a Trusted Execution Environment (TEE). 🔐 End‑to‑end encryption keeps prompts and responses private. ✅ Cryptographic attestation proves you’re talking to genuine secure hardware. 🚫 Zero data retention – no logs, no training data.

Why it matters:.Most AI‑powered apps (calorie trackers, CRMs, study companions, relationship apps eyes 👀) send every prompt to OpenAI, exposing all user data. With Maple Proxy, that data never leaves a hardware‑isolated enclave. Here it powers @jandotai

This changes everything. Introducing Maple Proxy – a powerful bridge that lets you call Maple’s end‑to‑end‑encrypted LLMs with any OpenAI‑compatible client, without changing a line of code.

“The single biggest use case is surveillance and control. This is another reason why things like Maple are incredibly important.” - Odell

Zero Data Retention is included with every Maple account by default, even free users. Other AI apps charge high amounts for merely a business promise of ZDR. We built zero access from the foundation using verifiable encryption. Make your choice.

RT @anthonyronning_: Don't forget to turn this off. Or use @TryMapleAI where every chat is end to end encrypted via confidential computin….

Maple is now listed in the PrivacyPack app by @enteio. Follow the arrow to better privacy