Run your AI employees on Gemini
Bring your own Google AI key and Coordinator's agents run on Gemini — reading vendor invoices as the PDFs they actually are, on a frontier lineup priced like a budget one, billed by Google directly to you at zero markup.
Free. No card. About two minutes to set up.
A Gemini-powered employee, not a chatbot
An agent here holds roles exactly like a person does, and its ceiling is its roles' ceiling: whatever someone with those roles could do in Coordinator, you can hand over. Gemini decides how well the work gets done — not how far it reaches.
Good delegation ends in a draft invoice, not a chat log.
- Read a forwarded vendor PDF into a draft bill, validated before it touches the books
- Watch receivables aging and draft the chase
- Answer P&L and client-profitability questions from the live ledger
- Create, file and update activities across your pipelines and boards
- Work a queue — claim the next item, do it, hand off
- Run the check-in round-up and keep stages honest
- Log time as it works
- Sweep unbilled hours into a draft invoice
- Keep the time ledger clean enough to bill from
- Triage form submissions into records and the right pipeline
- Keep client records current as work happens
- Find anything — records, contacts, history — on request
- Prep today's meetings from the calendar
- Tend booking-link intake and meeting polls
- Keep resources and rooms straight
- Draft replies, follow-ups and summaries inside your team's threads
- Announce what needs announcing
- Turn a meeting thread into the follow-up tasks
Hold the Owner role. Message outside your organization before you deliberately climb the outbound ladder. Appear human in the audit trail — agent work is labeled agent work, always. Keep going after you hit pause. Everything else opens closed and is yours to grant, role by role.
What Gemini itself brings
Any provider on this site can power the delegation above. These are the reasons to pick Gemini in particular.
Gemini models your agents can use
Prices and lineup verified September 2, 2026. In the product, the picker is fetched live from Google on your own key — it never goes stale, and a custom model id always works if it validates.
$0.75/$3.75 per million tokens (promotional through the end of 2026), million-token context, native PDF — the Flash line is now Google's lead tier
$0.25/$1.50 — the budget tier for high-volume extraction, still with the full context window
$2/$12 — the Pro line currently ships as a preview; capable, but preview models can move under you
the 2.5 GA models remain active with no announced shutdown — teams standardised on them are not stranded
new 3.x models appear in the picker the moment your key can call them — fetched from Google with the exact capability check the validator uses
Media through Coordinator's pipeline: PDFs and images. The authority on current rates is Google's pricing page.
Gemini's API, as its users describe it
Recurring themes from public developer forums and reviews — the likes and the complaints — and how each plays out when the provider runs inside your own workspace.
- Routinely called the most generous free tier of any frontier API — no card, no expiry, full million-token context — which is why so many teams prototype on Gemini first.
- Gemini 3 Pro's late-2025 launch was a benchmark landslide, and its price-performance ($2/$12 per million tokens) keeps it in every serious comparison.
- Million-token context plus multimodal input on the cheap tiers is the most-cited practical draw.
The free tier's quotas were quietly slashed in December 2025, and 429 quota errors became the dominant developer complaint of 2026.
Here: Coordinator validates your key with a live call at connect time and errors name the real cause — and for production agent work, the paid tier is the honest baseline anyway (it also carries Google's stronger data-handling terms).
An August 2025 billing bug produced shock bills (reversed, and followed by proper spend caps in March 2026) — trust took a dent.
Here: Your key, your Google Cloud project, your spend cap — set the project cap Google now provides, and the agent's usage lands in your own billing console where you can see it.
Model churn is real: the 1.x family is gone, 2.5 models were deprecated while 3.x was still in preview, and some cutoffs landed before their announced dates.
Here: The model picker is fetched live from Google's API on your key with the exact capability check the validator uses — a retired model disappears from the list instead of failing an agent mid-task, and a custom id lets you move to a new generation the day it appears.
How Coordinator treats your Google key
The section above is Google's report card. This one is ours — the plumbing decisions that make handing an agent your API key boring instead of brave.
Your key is encrypted like every credential in Coordinator, and after connecting, nothing ever displays more than its last four characters.
Connecting is a live round-trip to Google — a key that doesn't actually work can't be saved, so failures happen in settings, not mid-task.
Zero markup, no metering, no shared key anywhere in the system. Every agent turn bills your own account at the vendor's own prices.
Google's own docs show the API key in the query string; Coordinator never does that. The key rides the x-goog-api-key header only, so it stays out of proxy logs, browser history and error reports.
Model choices come from Google's own API using the exact capability check that validated your key — so a model you can pick is a model your key can genuinely call.
From no key to a working agent in a few minutes
Create a key at Google AI Studio → Get API key. Enable billing on the project before sending real customer documents.
- 1
Sign in at aistudio.google.com with the Google account you want this billed to.
- 2
Click Get API key (top left), then Create API key.
- 3
Choose an existing Google Cloud project or let it create one. The key belongs to that project — that is where usage and billing appear.
- 4
Copy the key. It starts with AIza and, unlike the other providers, you can come back and read it again later.
- 5
Open the project's billing settings and enable billing to move off the free tier. This is not only about rate limits — see the privacy note below.
Key verified — a live call to Google confirmed it before anything was stored. Only the last four characters are kept visible.
Settings → AI: pick Google (Gemini), paste the key. A live call to Google verifies it before anything is stored.
Fetched live from your key — never a stale list. Custom model ids work too.
Choose a default model from your key's live catalogue — or set one per agent later.
There is a free tier — Flash and Flash-Lite models only since April 2026, with low per-minute limits and around a thousand requests a day; enough to try extraction on a few bills without a card, and Google may use free-tier data to improve its products. On the paid tier a Flash model reads a one-page bill for well under a US cent; Pro-tier models cost several times more. Check the pricing page for current rates.
PRIVACY: on Google's FREE tier, prompts and responses may be reviewed by humans and used to improve Google's products. Your bills carry vendor names, amounts and tax IDs. Enable billing on the project — the paid tier is not used for training — before you send anything real through it.
Every agent turn bills your Google account at Google's prices. No markup, no shared key, live validation at connect time.
Gemini + Coordinator, asked directly
Any current Gemini chat model your key can reach — 3.8 Flash down to the Flash-Lite tiers, plus the still-served 2.5 family. The picker fetches the live list from Google on your key, and a custom model id always saves if it validates.
Yes — natively, alongside images. Gemini, Claude and OpenAI all receive a PDF as the file it is; Grok, Venice, Sail and OpenRouter take images, and Meta's PDF support runs through an API our shared path doesn't use yet.
No. Coordinator is free; Google bills your key directly at their published prices.
Swap the brain, keep the team
Your agents aren't built out of Gemini — they're built out of roles, permissions and workflows. The model behind an agent is a setting: change it in a minute and the team, the history and the fences all stay put. Every provider we support gets the same treatment, and the same honest page:
Put Gemini on the payroll
Coordinator is free — the only bill is the one Google already sends you.
Free. No card. About two minutes to set up.