Panaxon AI · M.N.E.M.O.

Add an AI chat agent to your product, without building the AI backend.

Panaxon AI is an API for custom chat agents that answer from your own data: your help desk, a knowledge search on your site, or an assistant inside your web or desktop app. M.N.E.M.O. handles the data, memory and retrieval, so you don't build a RAG pipeline or run a vector database. It isn't another ChatGPT. It's the engine behind the AI chat in your product.

Two products, one platform

A ready-made chat agent, or just the retrieval.

Both run on the same memory and the same exact-recall engine, so you can start with either and add the other without moving any data. Building this yourself means ingestion and chunking, embeddings, a vector database or knowledge graph, retrieval tuning, memory and access control. Panaxon AI is that stack, as an API.

1 · AI Chat Agent

A custom chat agent for your product

Put an agent on your help desk, behind your site's knowledge search or inside your app. It answers from your data with the exact stored values, remembers the conversation, and follows role-based access. Call it through the REST API.

From: the Free plan, with 1 API key for a try-out. Paid plans get unlimited API keys.

Learn more →

2 · MNEMO RAG Agent

Exact retrieval for your own pipeline

Already building your own agent or workflow? Define an endpoint over the subjects you choose and call it with one stateless request, in place of a vector database or knowledge graph.

From: every plan. Free includes 1 endpoint to try it.

Learn more →

Built into both

Your agent can answer from knowledge other tenants share.

The Knowledge Sharing Network lets one tenant publish a subject and others subscribe to it, without anyone receiving a file. A subscribed subject shows up in your chat agent and in your MNEMO RAG Agent endpoints exactly like your own data, with no limit on which product can use it. Subscribe to a vendor's documentation, a partner's catalogue or a regulator's updates, and your agent answers from your data and theirs. Publishers stay the owner and can revoke access at any time.

Where it fits

What businesses build with it.

Illustrative scenarios, not customer stories.

A help desk and support agent

Answers customer questions from your manuals, policies and ticket history, and says 'not found' when no path resolves.

Knowledge search for your site or app

A natural-language search box, like a search engine for your own content, that returns the exact stored answer and where it came from.

An assistant inside your product

Lets your users ask about their own data in your web or desktop app. Totals, lookups and comparisons are computed by the memory, not guessed.

An internal assistant

HR, IT or sales questions answered from the right subjects only, with role-based access rules on Pro and above.

The memory underneath

Everything runs on the same memory: exact recall, not approximate similarity.

Retrieval-augmented generation just means retrieving context before generating an answer. Vector search — approximate nearest-neighbor lookup over dense embeddings — is the common way to do that retrieval, not a requirement of it. M.N.E.M.O. does RAG through OAT, a patent-pending deterministic index: a query resolves to the exact stored fact, not the closest match by distance. That doesn't eliminate every source of error an LLM can introduce, but it removes the biggest one — being handed the wrong context in the first place, and not knowing it.

Predictable latency

Indexed retrieval is designed to fetch data directly, avoiding the variable cost of scanning approximate matches.

Scale without drift

Structured, hierarchical indexing is built to hold up as memory volume grows, not just at small scale.

Exact recall

Deterministic lookups are intended to reduce the drift that comes from approximate similarity matching.

Start anywhere

One account. Your plan decides how much of the platform you use.

Every plan runs on the same memory. Moving up a plan switches on more of the platform, with nothing to migrate.

PlanWhat you can use
FreeThe AI Chat Agent in Dorothy, 1 MNEMO RAG Agent endpoint and 1 API key for a try-out, with a limited monthly allowance, and subscribing to knowledge shared with you
StandardEverything in Free, plus publishing and private sharing, unlimited API keys and MNEMO RAG Agent endpoints, your own model provider key, and the MCP connector and Claude Code Skill
ProEverything in Standard, plus multiple users, role-based access rules and higher request limits
EnterpriseEverything in Pro, with negotiated usage, more model choice and dedicated support

The console

Dorothy: the test bed and control panel for your agent.

Dorothy is Panaxon AI's own app on M.N.E.M.O. Use it to load data, set up subjects and access rules, and try the answers before you wire the API into your product. It's also a working example of what the API can power, not a limit on what you can build with it.

Try the AI Chat Agent: open Dorothy and chat as a guest, no sign-up. Try MNEMO RAG Agent: create a free account and set up your one endpoint in Dorothy.

Next step

Start with what you need

Create an account in Dorothy to get your API key and try the memory on your own documents, or contact Panaxon AI about the agent you want to build.