What Bees sends over the network when you run a local model

Originally posted on bees.bot.

Our latest film is called "Nothing leaves your computer". So I went through the Bees code and listed every request it makes to the internet, to check that claim.

The short answer: when you run a local model, your prompts, your files, your agents' work and your memories stay on your computer. The model runs in llama-server, which only listens on 127.0.0.1, so nothing outside your machine can reach it. No crash reporting, no analytics and no tracking code ship in Bees.

But four things still go out by default, and one of them surprised me. I'd rather tell you than have you find it in a packet log.

What goes out without you doing anything

• app.bees.bot/api/config. At every start, Bees asks our server how sign-in is set up. It sends no account details and no token, only what any web request carries: your IP address and a user agent. It happens even if you never sign in. There is no switch in the app. Setting the BEES_ACCOUNT_API_URL environment variable points it somewhere else.
• The update check. At launch and every 6 hours, Bees fetches bees-update.json from our GitHub releases on github.com. The request carries no data about you. Nothing downloads until you click "Update and restart". There is no switch to turn this check off yet.
• api.freellmapi.co. This is the one I didn't expect. Bees bundles a small server that powers the free cloud AI option. It starts at every launch, even if you only use a local model. Ten seconds after start, and then every 12 hours, it asks api.freellmapi.co for the latest list of free models. It sends only the version of the list it already has. It sends a key only if you have a paid FreeLLMAPI license. Today the only way to stop it is the CATALOG_SYNC_DISABLED=1 environment variable, which most people will never set. We are fixing this so the server only runs when you pick the free AI option.
• The first memory setup. Local memory is on by default, so agents can remember what they learned. The first time it starts, Bees installs it with uv. That downloads Python 3.12 from GitHub, the memory packages from PyPI (pypi.org), and a small embedding model, sentence-transformers/all-MiniLM-L6-v2, from Hugging Face. Your memories are not part of those requests. Memory reads and writes go only to your local model; the code says in a comment, "Never forward memories to a hosted model". You can turn memory off for a workspace in settings. I'm not yet sure whether the memory service checks Hugging Face for a newer version of that embedding model each time it starts. I'll check on a clean install and update this post.

Our help page on what leaves your computer mentions the sign-in check and the model download. It does not mention the FreeLLMAPI list or the memory install. That page gets fixed too.

What goes out only when you or an agent asks

These need an action first. Each one sends what the action needs and nothing else.

• Downloading a model. When you turn on the Qwen3 4B model, Bees downloads one file of about 2.5 GB from huggingface.co over https and checks it against a fixed checksum. You can also paste your own GGUF link.
• Web search and page reading. Every agent has web search and page reading tools, and there is no switch for them. A search sends the query to search.brave.com, html.duckduckgo.com, lite.duckduckgo.com and bing.com together. News search uses Google News RSS. Reading a page fetches that page. These fire only when an agent decides to look something up, and the search query is the only part of your work that leaves. If a DEEPSEEK_API_KEY is set in the environment Bees starts from, one more search tool appears, and it sends queries to api.deepseek.com.
• Ask Bees planning. When you ask Bees to plan a setup, it may search the public add-on registry at registry.modelcontextprotocol.io and read skill packs from github.com and raw.githubusercontent.com. The Skills screen uses the same calls.
• The Apps screen. Opening it fetches our app list from bees-bot.github.io.
• Add-ons. None are on by default. When you add one, its package may download from npm or PyPI the first time it runs, and the add-on then talks to whatever service it is for.
• Gmail, Calendar and Drive. These reach Google only after you connect them.
• Cloud models. If you pick an API key provider, or a subscription like Codex, your prompts go to that provider. That is the whole point of picking one, and it does not happen on a local model.
• A Bees Cloud team. If you sign in and join a team, Bees syncs with app.bees.bot every 15 seconds. On my own Mac, which is signed in, I counted about 20 open connections from Bees to that server.

What I checked and found off

The agent runtime inside Bees has telemetry. Bees starts it with DSH_TELEMETRY_DISABLED=1, and I traced that setting through the runtime code to the place where it turns telemetry off. Even when telemetry is on, it only uploads after you send feedback.

The app window can't load anything remote. Its content security policy allows only the app itself and 127.0.0.1. No remote fonts, scripts or images. Help links open in your own browser when you click them.

The memory service has tracing code, but it only exports if an endpoint is set, and Bees doesn't set one. The device id file Bees keeps is only used when you are signed in to a team.

What I haven't checked

This list comes from reading the code. I have not yet watched the network on a fresh install. On this Mac I could only look at the Bees that was already running, which is signed in to a team, so it doesn't show the clean first run.

Two things I couldn't settle from the code. The memory reranker points at Hugging Face, but I couldn't find where it stores its download, so I don't know if it fetches one. And one library in the memory service can fetch a price list from GitHub when it loads, but Bees uses a different path that shouldn't load it. Next I'll run a clean install with nothing but a local model, watch the first 15 minutes of traffic, and post what I see.

What changes

The FreeLLMAPI list should not run when you never chose free cloud models. The update check should have a switch. The help page should list the memory install. Those are the three fixes I'm making first.

If you find a request I missed, open an issue. Bees is free and open source, on Mac, Windows and Linux. Download it at bees.bot, and the code is at github.com/Bees-bot/bees-desktop.