Gemini is HolaClaw's budget-friendly pick. Made by Google, it's fast, costs very little, and handles images and documents well — plus it can hold a lot of conversation in mind at once. There's even a free tier that's generous enough for a personal assistant. If you want a capable everyday brain without much spend, start here.
Fast, budget-friendly models with long context and strong multimodal skills.
- Authentication
- API key
- Cost
- $
- Recommended model
- Gemini 2.5 Flash
What you'll need
- A regular Google account — the same one you use for Gmail works fine.
- A few minutes in Google AI Studio, where you'll create your key.
- HolaClaw installed on your Mac.
You don't need to add a payment method to get started. Google's free tier lets your assistant chat right away, and you can upgrade later if you bump into its limits.
Get your API key
Gemini connects to HolaClaw with an API key — a long secret code that tells Google the requests are coming from your account. Think of it as a password your assistant uses on your behalf. Anyone who has it can run up usage on your account, so keep it private.
- Go to Google AI Studio and sign in with your Google account.
- Click Create API key.
- Copy the key and keep it somewhere safe. If you ever lose it, just come back and make a new one.
Tip — the free tier is real, but it has limits. Google's free tier is plenty for a personal assistant, but it caps how many messages you can send in a short window. If you chat heavily or run several assistants, you may hit those caps. Adding billing in Google AI Studio moves you to the paid tier and removes them — and even then, Gemini stays cheap.
Connect it to your assistant
You pick the provider while creating an assistant, in Step 2 · Model provider — the screen that says "Connect the brain behind your assistant."
- In HolaClaw, start the Create Assistant flow and go to Step 2.
- Choose Gemini in the provider sidebar.
- Paste your key into the Gemini API Key field. It's masked like a password, which is normal.
- Pick a model (more on that below) and continue with the rest of the setup.
As the app notes right below the key field: "Your credentials stay on your Mac — we never send them anywhere." The key is stored inside the assistant's private virtual machine on your Mac. HolaClaw's servers never see it, and Google bills you directly — HolaClaw adds nothing on top.
Choosing a model
HolaClaw offers six Gemini models. The first three are stable; the last three are previews.
| Model | Best for | Speed | Cost |
|---|---|---|---|
| Gemini 2.5 Flash (recommended) | Everyday chat, planning, images, documents | Fast | Very low |
| Gemini 2.5 Flash Lite | Light, high-volume chat on a tight budget | Fastest | Cheapest |
| Gemini 2.5 Pro | Harder reasoning and complex tasks | Slower | Higher |
| Gemini 3.1 Pro Preview | Trying Google's newest top-tier model | Slower | Higher |
| Gemini 3 Flash Preview | Trying the next-gen fast model | Fast | Low |
| Gemini 3.1 Flash Lite | Trying the next-gen lightweight model | Fastest | Low |
Gemini 2.5 Flash is the recommended default and the right pick for almost everyone — quick, inexpensive, and smart enough for a daily companion, a travel planner, or a study buddy. Flash Lite trims cost even further when you just want fast, frequent replies. Reach for 2.5 Pro when your assistant tackles genuinely hard problems — deep reasoning, intricate analysis — and you don't mind waiting a beat longer.
The three previews let you try Google's newest models early. They can be excellent, but Google may change or withdraw a preview at any time, so don't build your main assistant around one if you want stability.
Tip — Flash vs Pro, in plain terms. Think of Flash as a scooter: quick, cheap, perfect for getting around town all day. Pro is a car: more power for the long, demanding trips, but it costs more to run. Most days, the scooter is all you need.
What it costs
Google charges per token — roughly a short word or word-fragment of text. Every message you send and every reply your assistant writes uses some. On the free tier you pay nothing until you hit the rate limits; on the paid tier you pay only for what you use, with no monthly fee.
Gemini is one of the cheapest options here, so a chatty daily assistant on Flash typically costs very little — often a few cents a day, if anything, depending on which tier you're on. Pro costs more per message but is still modest. For current per-token prices, check Google's pricing from Google AI Studio — we don't quote numbers here because they change.
Troubleshooting
When something goes wrong with the connection to Google, HolaClaw shows a banner in the chat explaining what happened. Here's what each one means:
- "Gemini is out of credits" — your account can't cover usage right now. Check your billing settings in Google AI Studio and your assistant resumes on the next message.
- "Gemini API key is invalid, expired, or out of credits" — the key was deleted, mistyped, or the account can't pay. Create a fresh key in Google AI Studio and update it in the assistant's settings, or add billing.
- "Gemini rate limit or quota reached" — you've sent more requests than your tier allows in that window. This is the most common hiccup on the free tier. Wait a little and try again, or add billing in Google AI Studio to move to the paid tier and lift the limits.
- "Gemini is currently unreachable" — Google's servers are having a moment. Nothing to fix on your end; the next attempt goes through once their service recovers.
Still stuck? The HolaClaw Discord is a friendly place to ask.
Next steps
Not sure Gemini is the right brain for this assistant? Compare all the options in Choosing an AI provider — and remember the provider and model are set per-assistant at creation, so you can always spin up another assistant with a different one. If you'd like something warmer for long conversations, see Anthropic Claude; if you already pay for ChatGPT, OpenAI can reuse that. And if you'd rather try many models through a single key, the OpenRouter guide covers that.