> For the complete documentation index, see [llms.txt](https://docs.gpu-fi.uk/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.gpu-fi.uk/gpu-infrastructure.md).

# GPU Infrastructure

Describes the planned path from treasury growth to GPU infrastructure.

{% hint style="info" %}
Phase 1 has no physical GPUs. GPU infrastructure is a future development path, not a current capability.
{% endhint %}

GPUFi does not begin with a GPU fleet. It is building toward a GPU-powered ecosystem through progressive infrastructure development.

```
Trading activity
  ↓
Treasury accumulation
  ↓
First GPUs
  ↓
GPU rental
  ↓
External operators
  ↓
Inference API
  ↓
Compute-driven ecosystem revenue
```

In Phase 1, revenue and initial mining-pool funding come from trading tax. The treasury accumulates through its 30% creator-tax allocation. Physical infrastructure depends on sufficient treasury accumulation and later ecosystem activity.

### Infrastructure progression

| Phase   | Status and focus                                                     |
| ------- | -------------------------------------------------------------------- |
| Phase 1 | Trading-funded ecosystem. No physical GPUs.                          |
| Phase 2 | Treasury-funded first GPUs are planned.                              |
| Phase 3 | External GPU operators and permanent admin renouncement are planned. |
| Phase 4 | An inference API and developer compute demand are planned.           |

The roadmap describes intended future stages. It does not claim that later infrastructure is already deployed or guaranteed.

### Explore the infrastructure layer

{% content-ref url="/spaces/kkr7s1yzTsFyOUiijXzU/pages/ADvULPxMZNtHWoCSkIKt" %}
[Hardware Roadmap](/gpu-infrastructure/hardware-roadmap.md)
{% endcontent-ref %}

How GPUFi intends to move from no physical GPUs to an operating GPU infrastructure layer.

{% content-ref url="/spaces/kkr7s1yzTsFyOUiijXzU/pages/icuYlcRXPYF80kGElqFJ" %}
[GPU Operators](/gpu-infrastructure/gpu-operators.md)
{% endcontent-ref %}

How third-party GPU contributors are intended to participate in Phase 3.

{% content-ref url="/spaces/kkr7s1yzTsFyOUiijXzU/pages/wpuVSBuXOgnvn3iX4e5v" %}
[Inference API](/gpu-infrastructure/inference-api.md)
{% endcontent-ref %}

How developer-facing compute usage is intended to become another ecosystem revenue source in Phase 4.

Continue to [Hardware Roadmap](/gpu-infrastructure/hardware-roadmap.md) for the physical infrastructure path.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.gpu-fi.uk/gpu-infrastructure.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.