# AI use cases for Free Claude Code (Alishahryar1/free-claude-code)

Repo: [Alishahryar1/free-claude-code](https://github.com/Alishahryar1/free-claude-code) | Use cases: 4 | Page: https://dedupedai.com/use-cases/alishahryar1-free-claude-code/

FCC is an open-source Python proxy that lets coding agents such as Claude Code and Codex use free, paid, and local models from many providers through one Admin UI.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Run Claude Code on free models through FCC: https://dedupedai.com/use-cases/alishahryar1-free-claude-code/run-claude-code-on-free-models/
- Keep a coding session running during provider outages: https://dedupedai.com/use-cases/alishahryar1-free-claude-code/keep-coding-through-provider-outages/
- Run a Codex session on a project folder in the browser: https://dedupedai.com/use-cases/alishahryar1-free-claude-code/run-codex-sessions-in-the-browser/
- Route Claude model tiers to different models: https://dedupedai.com/use-cases/alishahryar1-free-claude-code/route-opus-and-haiku-to-different-models/

## How to run Claude Code on free models through FCC with Free Claude Code

Repo: [Alishahryar1/free-claude-code](https://github.com/Alishahryar1/free-claude-code) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Point Claude Code at a free NVIDIA NIM model and code from your terminal in about 30 minutes.

**Who it is for:** A developer who uses Claude Code daily and wants to cut model costs while keeping the same terminal workflow.

**The problem:** Claude Code normally needs a paid model connection, which makes long coding sessions expensive to try.

**The outcome:** Claude Code runs in the terminal and sends its requests to a model you chose in the FCC Admin UI.

### Steps

1. Run `curl -fsSL "https://raw.githubusercontent.com/Alishahryar1/free-claude-code/main/scripts/install.sh" | sh` on macOS or Linux, or the PowerShell command from the README on Windows, and choose Claude Code when prompted.
2. Run `fcc-server` on Linux, or open Free Claude Code from the desktop or Start menu on Windows or macOS.
3. Open the Admin UI URL from the server log, open NVIDIA NIM under Providers, click Add key, paste your key, and click Save.
4. Leave MODEL on the default or search the dropdown, then click Apply.
5. Run `fcc-claude` and select an FCC model from Claude Code's native /model picker.

### Prompt to paste into your agent

```text
Using the repo Alishahryar1/free-claude-code, set up FCC with a NVIDIA NIM key stored as <NVIDIA_NIM_API_KEY>, then start `fcc-claude`. Expected output: Claude Code opens and lists an FCC model in /model. Verify that a simple prompt returns a response before reporting success.
```

### Caveat

Free-tier availability and limits are set by each provider and may change, as the README notes.

Risk, remote script: The install step pipes a script from GitHub directly into sh, so review it first if your policy requires it.

Risk, credentials: Your provider API key is stored in FCC, so use a key you are willing to keep on this machine.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/alishahryar1-free-claude-code/run-claude-code-on-free-models/

## How to keep a coding session running during provider outages with Free Claude Code

Repo: [Alishahryar1/free-claude-code](https://github.com/Alishahryar1/free-claude-code) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Set a fallback model list so FCC moves to your next model when a provider fails.

**Who it is for:** A developer in the middle of a long coding task who cannot afford a stalled turn when one provider goes down.

**The problem:** When a model provider has an outage or errors out, the coding agent stops and the turn has to be restarted by hand.

**The outcome:** After retries are used up, FCC tries the next configured model, and the turn continues across every connected client.

### Steps

1. Install FCC with the install command from the README and configure at least two providers in Providers → Admin UI.
2. Open Model Config in the Admin UI and add an ordered Fallback Models list.
3. Click Apply to save the model selection.
4. Run `fcc-codex` or `fcc-claude` and work as normal.
5. Check the FCC server log after an outage to confirm which model handled the retry.

### Prompt to paste into your agent

```text
In the repo Alishahryar1/free-claude-code, configure two providers and an ordered Fallback Models list in Model Config, then start `fcc-codex`. Expected output: a coding turn completes even when the first model is unavailable. Verify by checking the server log for the fallback model name.
```

### Caveat

The README warns that a failed request may reach and consume usage from more than one provider before it succeeds.

Risk, credentials: Each provider you add needs its own API key, and every key stays stored in FCC.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/alishahryar1-free-claude-code/keep-coding-through-provider-outages/

## How to run a Codex session on a project folder in the browser with Free Claude Code

Repo: [Alishahryar1/free-claude-code](https://github.com/Alishahryar1/free-claude-code) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Pick a folder in FCC's browser UI and run Codex there, switching models mid-session.

**Who it is for:** A developer who wants to start a coding session from a browser tab, not a terminal.

**The problem:** Setting up a terminal coding agent for every project folder is slow when you only want a quick session.

**The outcome:** A native Codex session runs against the folder you chose, with real-time and background support.

### Steps

1. Run `curl -fsSL "https://raw.githubusercontent.com/Alishahryar1/free-claude-code/main/scripts/install.sh" | sh` on macOS or Linux, or the PowerShell command from the README on Windows, and choose Codex when prompted.
2. Start FCC with `fcc-server` on Linux, or open Free Claude Code on Windows or macOS.
3. Open the Admin UI URL from the server log and configure at least one provider.
4. Choose a folder and start a Codex session in the browser UI.
5. Switch providers or models inside the same session when needed.

### Prompt to paste into your agent

```text
Using the repo Alishahryar1/free-claude-code, start FCC, pick the folder <PROJECT_PATH> in the browser UI, and open a native Codex session. Expected output: Codex answers a question about the folder contents. Verify that the answer references files from <PROJECT_PATH>.
```

### Caveat

The README says switching harnesses inside one session is coming soon, so only provider and model switching works today.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/alishahryar1-free-claude-code/run-codex-sessions-in-the-browser/

## How to route Claude model tiers to different models with Free Claude Code

Repo: [Alishahryar1/free-claude-code](https://github.com/Alishahryar1/free-claude-code) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README

Send Opus, Sonnet, and Haiku requests to separate FCC models to balance cost and quality.

**Who it is for:** A developer who wants heavy reasoning on a strong model and quick edits on a cheaper one.

**The problem:** Claude Code expects its model tiers to map to one connection, which limits how much you can save per task.

**The outcome:** Each Claude model tier is sent to its own configured model, while tools, streaming, and images keep working.

### Steps

1. Install FCC with the install command from the README and choose Claude Code when prompted.
2. Start FCC with `fcc-server` on Linux, or open Free Claude Code on Windows or macOS.
3. In the Admin UI, configure the providers you want to use and select a model for each tier in Model Config.
4. Click Apply to save the routing.
5. Run `fcc-claude` and check the /model picker to confirm each tier.

### Prompt to paste into your agent

```text
In the repo Alishahryar1/free-claude-code, map the Opus, Sonnet, and Haiku tiers to different FCC models and start `fcc-claude`. Expected output: each tier shows its assigned model in the /model picker. Verify by running one short prompt on each tier and checking the server log.
```

### Caveat

The README says tiers are routed independently only with compatible models, so check each model's support before mapping it.

Risk, credentials: Each provider used for a tier needs its own key stored in FCC.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/alishahryar1-free-claude-code/route-opus-and-haiku-to-different-models/