Service eligibility and regional restrictions
PuppyIP serves only compliant overseas businesses and their authorized personnel. Proxy services are not available in mainland China. The service may only be used for lawful business activities outside mainland China. Use of this service within mainland China is prohibited.
Hosting a proxy IP or server overseas does not change these restrictions. The service must not be provided to end users in mainland China through relaying, forwarding, sharing or resale. Before use, read the Terms of Service.
Key Takeaways
- Hydrogen is generally available on Copilot Pro, Pro+, Max, Business, and Enterprise, with a gradual rollout across 11 listed client surfaces.
- Business and Enterprise administrators can manage model access through policy; default availability and model-specific settings can affect what members see.
- Per 1 million tokens, Hydrogen costs $1.50 for ordinary input, $0.075 for cached input, and $7.50 for output. One AI credit equals $0.01.
- The support matrix has no Hydrogen IDE minimum versions and does not list Hydrogen for Auto model selection or extended capabilities; do not assume 1 million-token context or configurable reasoning is supported.
Confirm your Copilot plan
On November 13, 2026, GitHub announced Gemini 4 Hydrogen, Google’s latest Flash model, as generally available in GitHub Copilot. The current supported-model matrix lists Copilot Pro, Pro+, Max, Business, and Enterprise. It does not list Free or Student as Hydrogen-supported plans; the setup documentation says those plans have access to models through Auto selection only, where Hydrogen is not currently listed. Individual paid plans do not need a separate model policy set up by an administrator.
The announcement names 11 client surfaces: Visual Studio Code, Visual Studio, Copilot CLI, Copilot coding agent, GitHub Copilot app, github.com, GitHub Mobile on iOS and Android, JetBrains IDEs, Xcode, Eclipse, and Copilot SDK. The support matrix also lists Hydrogen for its seven desktop or web clients: GitHub.com, CLI, Visual Studio Code, Visual Studio, Eclipse, Xcode, and JetBrains.
If the model is missing, check rollout and policy
GitHub says the launch is rolling out gradually. If you have a listed paid plan but cannot find Hydrogen in a listed client, your account may not have reached the rollout yet; the announcement advises checking again later. Public documentation does not give an account-by-account completion date, so another user’s model picker cannot confirm your own access.
Business and Enterprise administrators can manage model access in Copilot settings. New models inherit the organization or enterprise’s Default availability policy unless an administrator has turned that default off or explicitly disabled Hydrogen. An enterprise that restricts models to data-resident or FedRAMP-compliant options may also limit access. The available documentation does not establish Hydrogen’s compliance status for any particular organization. If you are an organization member, ask your administrator to check both the global default and any Hydrogen-specific setting.
Do not guess a minimum version or extra mode
GitHub’s support matrix currently marks the minimum versions for Hydrogen as TBD in Visual Studio Code, Visual Studio, JetBrains IDEs, Xcode, and Eclipse. TBD is not a version requirement, and it does not confirm support in every older release. Check the matrix for updates instead of reusing a minimum version published for another Gemini model.
Hydrogen is also absent from the current tables for Auto model selection and extended capabilities. Do not plan on Auto routing, a 1 million-token context window, or configurable reasoning as confirmed Hydrogen features. This guide covers availability and billing in Copilot; it does not make a speed, quality, or account-quota guarantee.
Estimate usage from the token rates
Tokens are units used to measure content processed by a model: input is sent to the model, output is generated by it, and cached input is reused context.
GitHub’s Copilot pricing table lists Hydrogen at $1.50 per 1 million ordinary input tokens, $0.075 per 1 million cached input tokens, and $7.50 per 1 million output tokens. The cached rate applies to tokens that billing recognizes as cached input. Your cost depends on the number of tokens in each category, not simply on how many prompts you send. The current table does not give Hydrogen a promotional end date, so do not copy another Gemini model’s promotion deadline.
For example, suppose one or more interactions total 100,000 ordinary input tokens and 20,000 output tokens, with no cached input. Input would cost about $0.15 and output about $0.15, for $0.30 or 30 AI credits. This is only a hypothetical rate calculation, not a measured run or a user’s extra bill. Actual token use and whether it exceeds a plan allowance depend on the account and workload.
Check credits, budgets, and legacy annual plans
One GitHub AI credit is worth $0.01. Included credit allowances vary across individual plans; Business and Enterprise per-user allowances are pooled at the relevant billing entity. Paid usage starts only after included allowances are exceeded and is governed by the account or organization’s budget settings. A listed token rate does not mean every interaction automatically creates an extra charge on top of a plan.
People who bought Pro, Pro+, or Max through an iOS or Android in-app purchase cannot buy extra AI credits through the App Store or Google Play. If their GitHub account has a payment method on file, however, paid usage beyond included credits can still be billed to that method according to budget settings.
To stop usage beyond the plan, set the Bundled AI credits budget to $0 and enable “Stop usage when budget limit is reached.”
Existing annual Pro or Pro+ subscribers who remain on legacy premium request-based billing do not receive new models or features. Their old request multipliers apply only to that legacy system, not to new usage-based billing.
Sources
- GitHub Changelog: Gemini 4 Hydrogen in GitHub Copilot
- GitHub Docs: Models and pricing for GitHub Copilot
- GitHub Docs: Supported AI models in GitHub Copilot
- GitHub Docs: Configuring access to AI models in GitHub Copilot
- GitHub Docs: Model multipliers for annual plans on request-based billing (legacy)
Frequently Asked Questions
Are code completions and next edit suggestions charged at these rates?
GitHub says code completions and next edit suggestions are not billed in AI credits. They remain unlimited on all paid Copilot plans and keep their existing counting mechanism. This does not establish that Hydrogen chat or agent interactions are free.
Does Hydrogen in Copilot give me access to the Google Gemini API?
That does not follow from the Copilot listing. The cited material establishes model availability and billing inside GitHub Copilot, but does not confirm access to the Gemini app, an Ultra plan, direct API access, or any Google API model ID.