### Install ComfyUI Dependencies and Run

Source: https://replicate.com/docs/guides/extend/comfyui.md

This snippet shows how to clone the ComfyUI repository, set up a virtual environment, install Python dependencies, and start the ComfyUI application.

```bash
git clone https://github.com/comfyanonymous/ComfyUI
# Optional: You might want to use a virtual environment
cd ComfyUI
python -m venv venv
source venv/bin/activate
# Install ComfyUI dependencies
pip install -r requirements.txt
```

```bash
python main.py
```

--------------------------------

### Setup Replicate Environment

Source: https://replicate.com/docs/guides/extend/comfyui.md

Install the necessary Python client library and configure the API token environment variable required for authentication.

```bash
pip install replicate
export REPLICATE_API_TOKEN=r8-*********************************
```

--------------------------------

### Install Replicate Python Client

Source: https://replicate.com/docs/guides/push-a-model

Install the Replicate Python client library using pip.

```bash
pip install replicate

```

--------------------------------

### Install and Run Vercel CLI

Source: https://replicate.com/docs/get-started/nextjs

Install the Vercel CLI globally and run it to initiate the deployment process, including login and app creation.

```bash
npx vercel
```

--------------------------------

### Install the Python Library

Source: https://replicate.com/docs/get-started/python.md

Install the official Replicate Python client library using pip.

```APIDOC
## Install the Python Library

### Description
Install the open-source Python client for the Replicate API using pip.

### Method
`pip`

### Endpoint
N/A

### Parameters
None

### Request Example
```plaintext
pip install replicate
```

### Response
N/A
```

--------------------------------

### Initialize Node.js Project and Install Replicate Client

Source: https://replicate.com/docs/get-started/nodejs.md

Manually set up a new Node.js project by creating a directory, initializing npm, setting the module type, and installing the official Replicate JavaScript client.

```shell
mkdir my-replicate-app
cd my-replicate-app
npm init -y
npm pkg set type=module
npm install replicate
```

--------------------------------

### Install Replicate Python Client

Source: https://replicate.com/docs/guides/run/upscale-images-with-ai-models.md

Commands to install the official Replicate Python package and configure the API token environment variable.

```bash
pip install replicate
export REPLICATE_API_TOKEN=<your-api-token>
```

--------------------------------

### List Examples for a Replicate Model

Source: https://replicate.com/docs/reference/http.md

Retrieves a list of example predictions made using a specific model. These examples are saved by the model author to illustrate the model's capabilities. The response is paginated.

```shell
curl -s \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  https://api.replicate.com/v1/models/replicate/hello-world/examples
```

--------------------------------

### Start Local Replicate MCP Server

Source: https://replicate.com/docs/reference/mcp.md

Use this command to start the local MCP server. It downloads and executes the replicate-mcp npm package. Press Ctrl+c to stop the server.

```plaintext
npx -y replicate-mcp
```

--------------------------------

### Install Cog

Source: https://replicate.com/docs/guides/push-a-model

Installs the Cog tool to your system. Ensure you have curl and appropriate permissions.

```bash
sudo curl -o /usr/local/bin/cog -L https://github.com/replicate/cog/releases/latest/download/cog_`uname -s`_`uname -m`
sudo chmod +x /usr/local/bin/cog
```

--------------------------------

### Install Cog on Linux Instance

Source: https://replicate.com/docs/guides/build/get-a-gpu-on-lambda-labs.md

Downloads and installs the latest version of the Cog binary to the system path and grants it execution permissions.

```sh
sudo curl -o /usr/local/bin/cog -L https://github.com/replicate/cog/releases/latest/download/cog_`uname -s`_`uname -m`
sudo chmod +x /usr/local/bin/cog
```

--------------------------------

### Install Deno

Source: https://replicate.com/docs/reference/mcp.md

This command installs Deno, a JavaScript and TypeScript runtime, which is required for executing code in the sandboxed environment when using Code mode.

```bash
curl -fsSL https://deno.land/install.sh | sh
```

--------------------------------

### Install Replicate Python Library

Source: https://replicate.com/docs/get-started/python.md

Installs the open-source Python client for the Replicate API using pip. This is the first step to programmatically interact with Replicate models.

```plaintext
pip install replicate
```

--------------------------------

### GET /v1/hardware

Source: https://replicate.com/docs/reference/http.md

Retrieve a list of available hardware configurations for running models.

```APIDOC
## GET /v1/hardware

### Description
Returns a list of available hardware objects that can be used for model execution.

### Method
GET

### Endpoint
https://api.replicate.com/v1/hardware

### Response
#### Success Response (200)
- **name** (string) - The display name of the hardware.
- **sku** (string) - The unique identifier/SKU for the hardware.

#### Response Example
[
  {"name": "CPU", "sku": "cpu"},
  {"name": "Nvidia T4 GPU", "sku": "gpu-t4"}
]
```

--------------------------------

### Copy Environment Variables Example (Shell)

Source: https://replicate.com/docs/guides/extend/openai-realtime.md

Copies the example environment variables file to a new file named .dev.vars. This is a common practice to set up local development environment variables based on a template.

```shell
cp .dev.vars.example .dev.vars
```

--------------------------------

### GET /v1/hardware

Source: https://replicate.com/docs/topics/models/create-a-model.md

Retrieves a list of available hardware configurations for running models.

```APIDOC
## GET /v1/hardware

### Description
Lists all available hardware types that can be used to run models on Replicate.

### Method
GET

### Endpoint
https://api.replicate.com/v1/hardware

### Response
#### Success Response (200)
- **name** (string) - The display name of the hardware.
- **sku** (string) - The SKU identifier for the hardware.

#### Response Example
[
  { "name": "Nvidia T4 GPU", "sku": "gpu-t4" },
  { "name": "Nvidia A40 (Large) GPU", "sku": "gpu-a40-large" }
]
```

--------------------------------

### Wrangler Dev Output Example (Plain Text)

Source: https://replicate.com/docs/guides/extend/openai-realtime.md

Example output from the 'npm run dev' command, showing the status of the local Cloudflare worker, available bindings (secrets), and the local server URL. It also provides interactive options for managing the development server.

```plaintext
> openai-realtime@0.0.0 dev
> wrangler dev
Your worker has access to the following bindings:
- Vars:
  - OPENAI_API_KEY: "(hidden)"
  - REPLICATE_API_TOKEN: "(hidden)"
⎔ Starting local server...
[wrangler:inf] Ready on http://localhost:3000
╭───────────────────────────╮
│  [b] open a browser       │
│  [d] open devtools        │
│  [l] turn off local mode  │
│  [c] clear console        │
│  [x] to exit              │
╰───────────────────────────╯
```

--------------------------------

### Initialize Vinext Project

Source: https://replicate.com/docs/guides/run/vinext.md

Commands to clone the starter repository and install the necessary project dependencies.

```shell
git clone https://github.com/replicate/getting-started-vinext
cd getting-started-vinext
npm install
```

--------------------------------

### GET /v1/models/{model_owner}/{model_name}/examples

Source: https://replicate.com/docs/reference/http.md

Lists example predictions made using a specific model. These examples illustrate the model's capabilities and are saved by the model author.

```APIDOC
## GET /v1/models/{model_owner}/{model_name}/examples

### Description
Lists example predictions made using the model. These are predictions that were saved by the model author as illustrative examples of the model’s capabilities. If you want all the examples for a model, use this operation. If you just want the model’s default example, you can use the `models.get` operation instead.

### Method
GET

### Endpoint
https://api.replicate.com/v1/models/{model_owner}/{model_name}/examples

### Parameters
#### Path Parameters
- **model_owner** (string) - Required - The name of the user or organization that owns the model.
- **model_name** (string) - Required - The name of the model.

### Request Example
```shell
curl -s \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  https://api.replicate.com/v1/models/replicate/hello-world/examples
```

### Response
#### Success Response (200)
A pagination object containing a list of example predictions.

#### Response Example
```json
{
  "next": "https://api.replicate.com/v1/models/replicate/hello-world/examples?cursor=...",
  "previous": "https://api.replicate.com/v1/models/replicate/hello-world/examples?cursor=...",
  "results": [...] 
}
```
```

--------------------------------

### Define Python Dependencies in cog.yaml

Source: https://replicate.com/docs/guides/push-a-model

Specifies the Python version and packages to be installed in the Docker image. Example uses PyTorch 2.3.1.

```yaml
build:
  python_version: "3.12"
  python_packages:
    - "torch==2.3.1"
```

--------------------------------

### Configure Webhook Events Filter (JSON)

Source: https://replicate.com/docs/topics/webhooks/receive-webhook.md

Example JSON payload to configure Replicate to send webhooks only for 'start' and 'completed' events.

```json
{
  "input": {
    "text": "Alice"
  },
  "webhook": "https://example.com/my-webhook",
  "webhook_events_filter": ["start", "completed"]
}
```

--------------------------------

### Stream File Data from Next.js Handler

Source: https://replicate.com/docs/how-does-replicate-work

Implement a Next.js API route to stream file data returned from a Replicate prediction. This example shows a basic GET handler.

```javascript
export async function GET(request: Request) {
  const output = await replicate.run(...);
  return new Response(output);
}

```

--------------------------------

### Initialize Project Directory

Source: https://replicate.com/docs/guides/run/discord-bot.md

Commands to create a new project directory and navigate into it using the command line.

```shell
mkdir flux-bot
cd flux-bot
```

--------------------------------

### Run Cog Model Prediction

Source: https://replicate.com/docs/guides/build/get-a-gpu-on-brev.md

Command to execute a Cog model and get predictions. It navigates to the model directory and uses `cog predict` with input arguments. The example shows predicting with the input text 'world'.

```plaintext
cd my-model
cog predict -i text="world"
```

--------------------------------

### Run ControlNet Tile Upscaler with JavaScript

Source: https://replicate.com/docs/guides/run/upscale-images-with-ai-models.md

Demonstrates how to initialize the Replicate client in Node.js, execute the high-resolution ControlNet tile model with specific input parameters, and save the resulting image to the local filesystem.

```bash
npm install replicate
export REPLICATE_API_TOKEN=<your-api-token>
```

```javascript
import Replicate from "replicate";
import fs from "fs/promises";
const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});
const output = await replicate.run(
  "batouresearch/high-resolution-controlnet-tile:latest",
  {
    input: {
      hdr: 0.2,
      image: "https://replicate.delivery/pbxt/K5vab12temDjc8jOnWFJ3p4RD7YWPQ3nuCyXpaRmN9yB8M1h/f84e7869-32ca-444b-a720-19e4325f4347.jpeg",
      steps: 20,
      prompt: "a woman wearing a colorful suit",
      scheduler: "DDIM",
      creativity: 0.6,
      guess_mode: false,
      resolution: 2560,
      resemblance: 0.4,
      guidance_scale: 5
    }
  }
);
fs.writeFileSync("output.png", output[0]);
```

--------------------------------

### Example Training Request Body (JSON)

Source: https://replicate.com/docs/reference/http.md

A sample JSON object representing the data required to start a new model training job. It includes the destination for the trained model, the input data for training, and an optional webhook URL.

```json
{
  "destination": "{new_owner}/{new_name}",
  "input": {
    "train_data": "https://example.com/my-input-images.zip"
  },
  "webhook": "https://example.com/my-webhook"
}
```

--------------------------------

### Get Model Details using cURL

Source: https://replicate.com/docs/reference/http.md

This example shows how to retrieve information about a specific model using a cURL command. It requires the model owner and model name in the URL and uses the API token for authentication. The response is a JSON object containing model details.

```shell
curl -s \n  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \n  https://api.replicate.com/v1/models/replicate/hello-world
```

--------------------------------

### Initialize Cog Project

Source: https://replicate.com/docs/guides/build/push-a-model.md

Navigate to your model directory and run the initialization command to generate the necessary configuration files.

```bash
cd path/to/your/model
cog init
```

--------------------------------

### Scaffold Node.js Project with create-replicate

Source: https://replicate.com/docs/get-started/nodejs.md

Use the `create-replicate` npm package to quickly scaffold a new Node.js project configured to run models on Replicate. This command automates project setup, including writing starter code and installing dependencies. You can optionally specify a model to use.

```shell
npx create-replicate
npx create-replicate --model black-forest-labs/flux-schnell
```

--------------------------------

### GET /v1/webhooks/default/secret

Source: https://replicate.com/docs/reference/http.md

Get the signing secret for the default webhook endpoint.

```APIDOC
## GET /v1/webhooks/default/secret

### Description
Retrieves the signing secret used to verify that incoming webhook requests originate from Replicate.

### Method
GET

### Endpoint
https://api.replicate.com/v1/webhooks/default/secret

### Response
#### Success Response (200)
- **key** (string) - The webhook signing secret.

#### Response Example
{
  "key": "whsec_..."
}
```

--------------------------------

### Clone Starter Project (Shell)

Source: https://replicate.com/docs/guides/extend/openai-realtime.md

Clones the starter repository for the OpenAI Realtime project from GitHub using Git. This command downloads the project files to your local machine, setting up the initial project structure.

```shell
git clone https://github.com/replicate/getting-started-openai-realtime openai-realtime-starter
cd openai-realtime-starter
```

--------------------------------

### Run Real-ESRGAN with Python

Source: https://replicate.com/docs/guides/run/upscale-images-with-ai-models.md

Example of using the Replicate Python client to run the Real-ESRGAN model. It takes an image URL as input and saves the resulting upscaled image to a local file.

```python
import replicate
output = replicate.run(
    "nightmareai/real-esrgan:latest",
    input={
        "image": "https://replicate.delivery/pbxt/Ing7Fa4YMk6YtcoG1YZnaK3UwbgDB5guRc5M2dEjV6ODNLMl/cat.jpg",
        "scale": 2,
        "face_enhance": False
    }
)
# Save the output image
with open('output.png', 'wb') as f:
    f.write(output[0].read())
```

--------------------------------

### Run Real-ESRGAN with Python

Source: https://replicate.com/docs/guides/run/upscale-images-with-ai-models.md

This snippet shows how to install the Replicate Python client, set your API token, and run the Real-ESRGAN model to enhance an image.

```APIDOC
## Install and Setup

### Description
Install the Replicate Python client and set your API token as an environment variable.

### Command
```bash
pip install replicate
export REPLICATE_API_TOKEN=<your-api-token>
```

## Run Real-ESRGAN

### Description
This example demonstrates how to use the `replicate.run` function to process an image with the `nightmareai/real-esrgan` model, specifying input parameters like image URL, scale, and face enhancement.

### Method
`replicate.run()`

### Endpoint
`"nightmareai/real-esrgan:latest"`

### Parameters
#### Input Parameters
- **image** (string) - Required - The URL of the image to enhance.
- **scale** (integer) - Optional - The scaling factor for the image.
- **face_enhance** (boolean) - Optional - Whether to enable face enhancement using GFPGAN.

### Request Example
```python
import replicate

output = replicate.run(
    "nightmareai/real-esrgan:latest",
    input={
        "image": "https://replicate.delivery/pbxt/Ing7Fa4YMk6YtcoG1YZnaK3UwbgDB5guRc5M2dEjV6ODNLMl/cat.jpg",
        "scale": 2,
        "face_enhance": False
    }
)
```

### Response
#### Success Response (200)
- **output** (list of file objects) - A list containing the enhanced image file(s).

### Response Example
```python
# Save the output image
with open('output.png', 'wb') as f:
    f.write(output[0].read())
```
```

--------------------------------

### Update Installed Agent Skills

Source: https://replicate.com/docs/reference/skills.md

Pulls the latest versions of all agent skills that have been previously installed.

```plaintext
npx skills update
```

--------------------------------

### Run Ultimate SD Upscale with Python

Source: https://replicate.com/docs/guides/run/upscale-images-with-ai-models.md

Shows how to use the Replicate Python client to run the Ultimate SD Upscale model. It includes the necessary installation steps and the model execution call with prompt and denoising configurations.

```bash
pip install replicate
export REPLICATE_API_TOKEN=<your-api-token>
```

```python
import replicate
output = replicate.run(
    "fewjative/ultimate-sd-upscale:latest",
    input={
        "image": "https://replicate.delivery/pbxt/KKOaCNy9baG5cUZfK3YvvePBqZeTxyvytifSo7pFOKLySQN2/ComfyUI_00004_.png",
        "steps": 20,
        "denoise": 0.4,
        "scheduler": "karras",
        "upscale_by": 2,
        "negative_prompt": "ugly, broken, weird",
        "positive_prompt": "a portrait photo of a cat in a green hoodie",
    }
)
```

--------------------------------

### Install Cog CLI

Source: https://replicate.com/docs/guides/build/optimize-models-with-pruna.md

Installs the Cog tool for containerizing and deploying machine learning models.

```bash
sh <(curl -fsSL https://cog.run/install.sh)
```

--------------------------------

### Handle File Outputs (Python)

Source: https://replicate.com/docs/get-started/python.md

Explains how to manage models that produce file outputs, such as images or audio. It shows how to save a single generated image and how to handle multiple file outputs.

```python
output = replicate.run(
    "black-forest-labs/flux-schnell",
    input={"prompt": "A majestic lion"}
)
# Save the generated image
with open('lion.png', 'wb') as f:
    f.write(output[0].read())
print("Image saved as lion.png")
# Handle multiple outputs
output = replicate.run(
    "black-forest-labs/flux-schnell",
    input={"prompt": "A majestic lion", "num_outputs": 2}
)
for idx, file_output in enumerate(output):
    with open(f'output_{idx}.png', 'wb') as f:
        f.write(file_output.read())
```

--------------------------------

### Install Brev CLI

Source: https://replicate.com/docs/guides/build/get-a-gpu-on-brev.md

Installs the Brev command-line interface using Homebrew for macOS or Linux systems.

```bash
brew install brevdev/homebrew-brev/brev
```

--------------------------------

### Initialize Cog Project

Source: https://replicate.com/docs/guides/build/optimize-models-with-pruna.md

Creates a new directory and initializes a Cog project structure.

```bash
mkdir my-pruna-model
cd my-pruna-model
cog init
```

--------------------------------

### Replicate Prediction Object Example (JSON)

Source: https://replicate.com/docs/topics/webhooks/receive-webhook.md

An example of the JSON object representing an unfinished prediction, as sent in webhook requests.

```json
{
  "id": "ufawqhfynnddngldkgtslldrkq",
  "version": "5c7d5dc6dd8bf75c1acaa8565735e7986bc5b66206b55cca93cb72c9bf15ccaa",
  "created_at": "2022-04-26T22:13:06.224088Z",
  "started_at": null,
  "completed_at": null,
  "status": "starting",
  "input": {
    "text": "Alice"
  },
  "output": null,
  "error": null,
  "logs": null,
  "metrics": {}
}
```

--------------------------------

### Run Model with Local File Input (Python)

Source: https://replicate.com/docs/get-started/python.md

Shows how to use a local file, like an image, as input to a Replicate model. This example uses the LLaVA vision model with a local image file.

```python
import replicate
image = open("my_fridge.jpg", "rb")
output = replicate.run(
    "yorickvp/llava-13b:a0fdc44e4f2e1f20f2bb4e27846899953ac8e66c5886c5878fa1d6b73ce009e5",
    input={
        "image": image,
        "prompt": "Here's what's in my fridge. What can I make for dinner tonight?"
    }
)
print(output)
# You have a well-stocked refrigerator filled with various fruits, vegetables, and ...
```

--------------------------------

### Run Real-ESRGAN with JavaScript

Source: https://replicate.com/docs/guides/run/upscale-images-with-ai-models.md

Demonstrates how to use the Replicate JavaScript client to perform image upscaling using the Real-ESRGAN model. It includes installation, environment configuration, and the API call to process an image URL.

```bash
npm install replicate
export REPLICATE_API_TOKEN=[your-api-token]
```

```javascript
import Replicate from "replicate";
const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});
const output = await replicate.run(
  "nightmareai/real-esrgan:latest",
  {
    input: {
      image: "https://replicate.delivery/pbxt/Ing7Fa4YMk6YtcoG1YZnaK3UwbgDB5guRc5M2dEjV6ODNLMl/cat.jpg",
      scale: 2,
      face_enhance: false
    }
  }
);
console.log(output);
```

--------------------------------

### Install and configure Replicate CLI

Source: https://replicate.com/docs/guides/run/make-art-with-stable-diffusion.md

Commands to install the Replicate CLI on macOS and set the required API token environment variable.

```shell
brew tap replicate/tap
brew install replicate
```

```plaintext
export REPLICATE_API_TOKEN=...
```

--------------------------------

### Create GitHub Repository and Push Code

Source: https://replicate.com/docs/get-started/nextjs

Use the GitHub CLI to create a new public repository and push your local code to it. Use the `--private` flag for private repositories.

```bash
gh repo create my-replicate-app --public --push --source=.
```

--------------------------------

### Run a Public Model with Text Prompt (Python)

Source: https://replicate.com/docs/get-started/python.md

Demonstrates how to run a public Replicate model, such as black-forest-labs/flux-schnell, using a text prompt and saving the generated image output.

```python
import replicate
output = replicate.run(
  "black-forest-labs/flux-schnell",
  input={"prompt": "an iguana on the beach, pointillism"}
)
# Save the generated image
with open('output.png', 'wb') as f:
    f.write(output[0].read())
print(f"Image saved as output.png")
```

--------------------------------

### Example secret list output

Source: https://replicate.com/docs/guides/cloudflare-image-cache

Example output from `npx wrangler secret list`, showing the names and types of secrets configured for the worker.

```json
[
  {
    "name": "CLOUDFLARE_ACCOUNT_ID",
    "type": "secret_text"
  },
  {
    "name": "CLOUDFLARE_API_TOKEN",
    "type": "secret_text"
  },
  {
    "name": "CLOUDFLARE_IMAGE_ACCOUNT_HASH",
    "type": "secret_text"
  },
  {
    "name": "REPLICATE_API_TOKEN",
    "type": "secret_text"
  }
]
```

--------------------------------

### Create a deployment using the Replicate API

Source: https://replicate.com/docs/topics/deployments/create-a-deployment.md

This snippet demonstrates how to create a new model deployment by sending a POST request to the Replicate API. It requires an API token for authentication and a JSON payload specifying the model, version, and hardware configuration.

```shell
curl -s \
  -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
        "name": "my-app-image-generator",
        "model": "some-org/some-model",
        "version": "da77bc59ee60423279fd632efb4795ab731d9e3ca9705ef3341091fb989b7eaf",
        "hardware": "gpu-t4",
        "min_instances": 1,
        "max_instances": 5
      }' \
  https://api.replicate.com/v1/deployments
```

--------------------------------

### Install Replicate npm Package

Source: https://replicate.com/docs/guides/extend/cloudflare-image-cache.md

Installs the Replicate npm package, which is necessary for interacting with Replicate's API from your Node.js or JavaScript environment.

```shell
npm install replicate
```