$300 in Google Cloud free credits is available to new customers, and you get 91 days to use it across most GCP services. If you want to spin up a real project (not just a toy demo), this is one of the better starter deals in cloud.
Startup builders trying to stretch runway, developers evaluating Vertex AI, and students doing research workloads all end up here for the same reason. You can test Compute Engine, BigQuery, Cloud Storage, and a lot of the core stack without paying, as long as you don’t upgrade.
This guide covers Google Cloud Free Trial eligibility, the exact signup steps, the restrictions that trip people up, and a few practical ways to make the credits go further.
Program at a Glance
| Provider | |
| Credit Amount | $300 in Google Cloud “Welcome” credits |
| Duration | 91 days from signup |
| Eligibility | New Google Cloud customers only; one trial per person/org |
| Credit Card Required? | Yes, for identity verification (temporary $0–$1 hold) |
| Difficulty | Moderate; card verification and banking rules can block signup |
| Best For | Prototyping on GCP, trying Vertex AI, testing always-free services |
| Official Page | Google Program Page |
What You Actually Get
The Google Cloud Free Trial gives new customers $300 in credits to explore “virtually all” Google Cloud Platform services for 91 days. In practice, people use it for things like Compute Engine VMs, BigQuery queries and storage, Cloud Storage buckets, and Vertex AI features such as Vertex AI Studio (prompting, tuning, and testing Gemini models). It also pairs nicely with Google’s always-free services, which continue even after the trial ends and do not consume the $300.
What’s the real-world value? $300 is enough to run a small dev environment, test CI/CD, stand up APIs on Cloud Run/Cloud Functions, and explore data workflows in BigQuery without rushing on day one. It is not “free GPU time” on GCP during the trial, though. That’s the catch most ML folks hit quickly.
Who Qualifies (and Who Doesn’t)
This free trial is for new Google Cloud customers, and Google is fairly strict about that definition. If you’re trying to grab a second $300 on the same identity or billing history, expect the application to fail or the account to be ineligible.
- You must be a new Google Cloud customer, and it’s limited to one free trial per person or organization.
- A Google account is required, and a personal Gmail account works fine.
- You need a valid credit or debit card for identity verification, and Google may place a temporary $0–$1 authorization hold.
- Your card has to allow automatic payments (some banks block recurring authorizations, which can break signup).
If you’ve previously been a paying customer of Google Cloud, Google Maps Platform, or Firebase, you don’t qualify for the free trial. Also, prepaid cards may not work, so don’t plan on that workaround.
How to Sign Up
Signup usually takes about 10 minutes if your card verification goes smoothly.
- Go to console.cloud.google.com or cloud.google.com/free.
- Sign in with any Google account (a personal Gmail account is fine).
- Click “Start free” or “Get started for free”.
- Select your country and accept the Terms of Service.
- Complete credit card verification by entering a valid credit or debit card. Google places a temporary $0–$1 USD authorization hold (not a charge) that is released within a few days to about a month depending on your bank; prepaid cards may not work, and some banks block automatic payments (notably in India due to RBI rules), which can cause signup failure.
Once you finish verification, your Free Trial billing account is created with $300 in Welcome credits. The 91-day clock starts immediately at signup, not when you deploy your first service.
What the Credits Cover
The $300 trial credits can be used across most Google Cloud services, and Vertex AI is explicitly in-scope. On top of that, Google offers “always-free” monthly usage limits for many AI and non-AI services, and those limits do not consume your $300 credit.
| Service / Feature | What It Does | Included? |
|---|---|---|
| Compute Engine | Virtual machines for backend services and batch jobs. | ✓ (with trial limits) |
| Vertex AI Studio | Prompt design, tuning, and testing with Gemini models. | ✓ (uses $300 credits) |
| BigQuery | Data warehouse for SQL analytics and ETL-style querying. | ✓ |
| Always-free AI APIs | Monthly free quotas for Vision, NLP, STT, Translation, and more. | ✓ (doesn’t use $300) |
Notable exclusions matter here: you can’t add GPUs to VM instances during the free trial, and GPUs on Vertex AI aren’t available until you upgrade to a paid account. You also can’t use Google Cloud Marketplace products during the trial.
Limitations to Know About
Every free program has catches. Google’s are clear once you know where to look, but they can surprise you if you’re trying to do “real” ML infrastructure on day one.
- You cannot add GPUs to Compute Engine VM instances until you upgrade to a paid account.
- Google Cloud Marketplace products are not available during the free trial.
- You can’t request a quota increase while you are on the free trial.
- Windows Server VM instances can’t be created during the free trial, and you’re capped at 8 concurrent Compute Engine CPU cores.
When credits run out or the 91 days expire, Google auto-closes the trial billing account if you don’t upgrade, and all running workloads are stopped (not deleted immediately). You get a 30-day grace period to upgrade and recover resources; after that, workloads are permanently deleted. If you do upgrade to paid, you keep any remaining credits until the 91-day expiration, restrictions lift immediately, and you only pay for usage beyond always-free limits.
Have Unused Google Credits?
Google credits have a habit of expiring while you’re busy shipping product. Startup programs and enterprise agreements can leave teams sitting on balances they’ll never burn down in time. If you’re in that situation, AI Credit Mart lets you sell unused credits at a steep discount instead of watching them go to zero. Honestly, it’s the cleanest option when finance is asking what happened to that big credit line item.
Need More Google Credits?
Once your $300 runs out, paying list price is not your only path. AI Credit Mart lists discounted Google credits from companies with surplus allocations that won’t be used before expiry. Discounts typically land around 30–70% below retail, which can buy you more time to validate a workload before committing to full spend.
Tips for Getting the Most Out of Your Credits
- Start with a plan for the 91-day clock, because it begins at signup, not first usage.
- Use Gemini API via Google AI Studio for the permanent free tier when it fits, since it’s separate from the $300 trial and doesn’t require a credit card.
- Be deliberate about Vertex AI versus AI Studio, because they are separate platforms with different billing, rate limits, and access patterns.
- Stick to always-free regions and limits where possible, like the always-free Compute Engine e2-micro VM in us-east1/us-west1/us-central1 and the always-free Cloud Storage regional quota in US regions.
- If you upgrade to paid, set budget alerts in the Billing console immediately, because Google Cloud doesn’t give you a default spending cap.
Frequently Asked Questions
They’re worth $300 in usage across most Google Cloud services for 91 days, including core GCP products like Compute Engine, BigQuery, Cloud Storage, and Vertex AI (for example, Vertex AI Studio with Gemini models). The practical value depends on what you run: you can build a real prototype, deploy an API, and test data workflows. The biggest caveat is that some high-demand resources are restricted on the trial (notably GPUs). If you need GPUs immediately, you’ll need a different path.
Yes, but it’s for identity verification only, with a temporary $0–$1 authorization hold.
The $300 Google Cloud Free Trial credits last 91 days from the day you sign up.
Yes. If you have Google credits you won’t use before they expire, you can list them on AI Credit Mart and sell them at up to 70% of face value. Companies regularly list surplus credits from startup programs and enterprise agreements.
AI Credit Mart has discounted Google credits available from companies with surplus allocations. Prices are typically 30-70% below retail.
If you don’t upgrade, your trial billing account auto-closes, workloads are stopped, and you get a 30-day grace period before resources are permanently deleted.
No. GPUs for Compute Engine VMs and GPUs on Vertex AI aren’t available during the free trial; they unlock only after you upgrade to a paid account.
Gemini API via Google AI Studio has its own permanent free tier that’s independent of the $300 trial. Vertex AI usage (like Vertex AI Studio with Gemini models) uses your GCP billing setup and can consume the $300 credits. Rate limits for the AI Studio free tier are roughly 5–15 requests per minute and up to about 1,000 requests per day depending on the model, and limits apply per Google Cloud project. Commercial use is allowed on that free tier.
$300 in credits plus always-free tiers is plenty to validate a real GCP setup fast. Sign up, keep an eye on the trial restrictions, and if you end up with surplus credits later, you’ve got a place to sell them.
Your AI credits are losing value every day
Join the marketplace and start trading unused credits today.