Google Gemini is now available as an agent in boostN — Mistral is next
You can now work with Google Gemini as an agent in boostN. If you have a Google subscription, you can run your AI work straight through our execution pipeline — no extra tooling, no wiring up your own API keys. That makes Gemini the next model you can pick alongside the agents boostN already supports. And it won’t be the last: the Mistral integration is already in the works and close to done.
What actually changed
- Google Gemini is live as an agent in boostN — selectable in the execution pipeline.
- Your Google subscription is enough: you work with the quota you already have at Google.
- Model choice, not model lock-in: Gemini sits next to the existing agents — you decide per task.
- Mistral is next: the integration is far along and ships in the coming days.
How it was before
Until now, agent work in boostN ran on a fixed set of models. If you held a subscription with a specific provider — Google, say — you couldn’t put that quota to use directly with us. Picking a model was less a per-task decision than a given. For anyone already living in the Google ecosystem, that meant a second subscription or a detour through third-party keys.
How it works now
1. Gemini is selectable as an agent. In boostN you start an agent and can now choose Google Gemini as the model. The agent plans the task, writes code, runs steps — same as always, just on Gemini.
2. Your Google subscription becomes the working basis. You don’t buy a new quota. If you have a Google subscription, that’s the quota you use for your work in boostN. It lowers the barrier for everyone already in the Google ecosystem.
3. One agent, one pipeline — different models. The core is our automatic execution pipeline: you kick off an agent, the pipeline calls the chosen model in the background and returns the result. Which model runs underneath is a setting, not a rebuild. This model openness builds on Google’s agentic ecosystem around Antigravity, which deliberately allows third-party models.
Why it matters
The point isn’t “one more model.” The point is that the model becomes a choice instead of a fixed setting. Rely on a single provider and you’re tied to its pricing and quality curve. A pipeline that knows several models lets you switch when another one is cheaper or better for a given task — without rebuilding your workflow.
To stay honest: Gemini is live, Mistral isn’t yet. We say “coming in the next few days” because the integration is far along — but until it ships, only what you can actually use counts. Gemini counts as of today.
What you can do now
If you have a Google subscription: pick Gemini as your agent in boostN and run your next task on it. You don’t need your own API key — your existing quota is enough.
If you’re waiting on Mistral: the integration is close to done. We’ll let you know the moment it’s live — until then, Gemini is the fresh option.
If you’re unsure which model fits: run the same task on two models and compare. That’s exactly what model choice is for — you don’t have to commit up front.
Learn more
- boostN — the web app at a glance
- What boostN can do — features
- Documentation & guides
- Google Antigravity: the agentic dev suite behind the model openness
Entdecke mehr
Headless without an API bill — how do you reach the best AI models for automation in 2026?
Provider comparison mid-2026: who has a headless mode, whose subscription still covers it — and why BYOK is the most stable foundation.
GlossarAider
Aider is an open-source AI pair programmer for the command line that edits code directly in the Git repository and commits every change automatically.
LexikonComparing AI models — who builds what and how to choose
The major model families in 2026 at a glance. Who builds Claude, GPT, Gemini, Llama, Mistral, DeepSeek, Qwen — and which model to pick when.