Using Brave Leo with Any LLM on the Planet

Try the Requesty Router and get $6 free credits 🔀

AI assistants have become a common part of modern web browsers. Brave has Leo, Edge has Copilot, Opera has Aria, Arc has Max… This new wave of browser-based AI raises many questions: How can we give users the freedom to pick any Large Language Model (LLM), while also safeguarding data privacy, preserving security, and keeping the browsing experience smooth?

Brave’s answer: Bring Your Own Model (BYOM). This optional new feature for Brave Leo lets you connect the browser’s AI interface to any LLM endpoint you choose. Rather than confine you to a single, proprietary provider, Brave hands you the keys: run the AI you want, at the endpoint you prefer—and keep full control of your data and usage.

With BYOM, requests go directly from your device to the model endpoint you configure. Brave doesn’t act as an intermediary or see that traffic. Your prompts, webpage content, or PDF text flow straight to the provider you designate. Brave just serves as the user interface in the browser. If you choose a high-performance cloud solution (like the Requesty router we’ll discuss below), you can get the best of all worlds: your model of choice, built-in analytics, fallback policies, robust cost controls, and easy switching whenever new models emerge.

Your model, your rules—with Requesty

One of the most powerful ways to leverage BYOM in Brave Leo is through Requesty, a universal LLM router that supports any model—from the latest Deepseek release, to Qwen by Alibaba, to Mistral, or just about any other you can think of. Requesty provides a single API endpoint to access multiple LLMs. You get logging, analytics, cost control, and the ability to automatically switch or “fall back” to another model if your primary one is overloaded or offline.

Here’s how to connect Brave Leo to Requesty, so you can start using any model in the world right in your browser:

1. Create a Requesty API key

   • Visit app.requesty.ai/router and sign up or log in.

   • In the dashboard, create / copy your API key. This single key is all you need to route requests to multiple providers.

2. Pick a model from Requesty

   • Browse the Model List. You’ll see cutting-edge LLMs like the newest Deepseek, Qwen from Alibaba, Mistral, or more specialized enterprise-focused models.

   • Each model has an identifier. Copy that ID (or keep it handy).

3. Configure Leo’s BYOM in Brave

   • Open Brave Settings → Leo.

   • Scroll to the Bring your own model section, click Add new model.

   • Label: Give the model any name you like (e.g., “Deepseek Latest” or “Alibaba Qwen”).

   • Model request name: Enter the model ID from Requesty (the one you saw in the list).

   • Server endpoint: Use https://router.requesty.ai/v1/chat/completions.

   • Authentication credentials: Paste the API key you got from Requesty.

4. Click “Add model”

   • Your new model will now appear in Leo’s model selection menu.

   • Select it. You’re done! Any queries you type into Leo will now route to your chosen LLM via Requesty.

Why use Requesty with Brave Leo?

• Limitless Model Choice
  Requesty aggregates multiple providers behind a single endpoint. That means you can switch from Qwen to the latest version of Deepseek or Mistral with just a change in your BYOM settings—no need to reconfigure everything.

• Fallback Policies
  If your go-to model is down or hits a rate limit, Requesty can automatically reroute your request to another model in seconds. This ensures you’re never stuck waiting for an LLM that’s not responding.

• Cost & Rate Limit Management
  Requesty offers a consolidated view of usage across different models. Set budgets and monitor your token usage in one centralized dashboard—rather than juggling multiple provider portals.

• Analytics & Logging
  Want deep insights into your AI usage? Requesty logs relevant request and response data so you can track performance, user patterns, and costs. At the same time, you can configure how much detail you want to store, so you stay in control of privacy.

• Security & Compliance
  Brave Leo and BYOM already keep your data out of Brave’s reach. With Requesty, you can also enforce organization-wide data policies, block certain types of content, or integrate advanced usage rules. Every request to the LLM is done over secure HTTPS, and you can even layer in additional safeguards via Requesty’s features.

• No Code Changes Required
  Because Brave Leo acts as the front end, and Requesty seamlessly handles routing, you don’t need to rewrite or patch any code in your apps or environment. Just configure BYOM and you’re set.

Using BYOM day to day

With Requesty as your BYOM endpoint, Brave Leo becomes an all-in-one AI console for your daily workflow. Summarize long articles, answer coding queries, or draft an email with the style of your favorite LLM—right from the browser sidebar. If a new model is announced tomorrow, just add it to Requesty, point Leo to that model ID, and keep working.

Best of all, your data is never sent to Brave’s servers. It’s a direct line from your machine to Requesty’s endpoint (which then talks to whichever model you’ve chosen). You remain fully in control of your browsing and LLM usage.

Wrapping up

Brave believes in giving you options. BYOM is about empowering users to run their AI their way—without vendor lock-in and without surrendering private data to a chain of unknown services.

Combining Brave Leo’s intuitive browser integration with Requesty’s universal LLM router is a powerful way to get the best of both worlds:

• Immediate AI assistance, inside your browser

• Freedom to use any model you prefer

• Consolidated analytics, fallback, and cost control

• No additional complexity or rewriting your environment

Ready to give it a try? Sign up for Requesty to grab your free API key, open Brave Leo settings, and bring your own model online. You’ll have a full roster of advanced AI tools—Deepseek, Qwen, Mistral, or countless others—right at your fingertips.

Enjoy the future of browser AI, your way. And stay tuned for more updates as Brave continues to expand on BYOM, ensuring that everyone can choose how they want AI to work for them.