Llama, Meta's family of open models, is one of the most widely used for internal assistants and document search. Connected to Mailpro, it no longer just answers: it prepares and sends.
Llama has been published by Meta, in Menlo Park, United States, since 2023. Its latest generations are multimodal and handle very long contexts; they can call tools.
Llama is distributed under Meta's community license, not a classic open source license: it allows commercial use, with conditions (notably for very large services). Read it before deploying.
In 2026, Meta launched a closed model, Muse Spark, alongside Llama. The Llama models already published remain available and widely installed in companies.
| Developer | Meta, Menlo Park (United States) |
|---|---|
| License | Llama community license (commercial use under conditions) |
| Common uses | Internal assistants, document search, models fine-tuned for a specific business |
| Tools | Tool calling supported |
| Note | Meta has also offered a closed model since 2026 |
Your team asks “who didn't open the last offer?” and Llama prepares the follow-up.
Llama finds the information in your internal documents and turns it into a message for your customers.
If you have fine-tuned Llama on your vocabulary, your emails benefit.
The same internal assistant answers questions and handles email marketing, with Mailpro's safeguards.
Run Llama with Ollama or LM Studio, or on the server that already hosts your internal assistant.
Use an interface that supports remote MCP servers: Open WebUI, LibreChat or LM Studio.
Add the Mailpro MCP server using the server address shown in your Mailpro account, on the AI Assistants (MCP) page, with OAuth sign-in (details: self-hosted AI).
In Mailpro, on the AI Assistants page, set how it sends: with approval or direct, with a daily cap.
Meta's community license is not a classic open source license: check that it suits your use and your country.
If your company has a Llama assistant in Open WebUI or LibreChat, adding Mailpro is just an admin setting.
An internal assistant used by the whole team deserves a cautious daily cap and approval for campaigns.
On the Mailpro side, the rules are the same for every assistant: by default, each send waits for your approval, or goes out directly within a daily cap; a log keeps track of everything, and one click suspends all sends. The details.
Llama is distributed under Meta's community license, which allows commercial use under conditions; it is not a classic open source license.
Yes, recent Llama models support tool calling; go through an interface that supports MCP servers, such as Open WebUI.
Yes, if its interface supports remote MCP servers with OAuth: Open WebUI and LibreChat do.
All questions: AI assistants FAQ.
Create a free Mailpro account, connect your assistant from the AI Assistants page of your account, and stay in control of every send.
