### Setup environment for video understanding

Source: https://docs.parasail.io/parasail-docs/guides/multi-modal

Install required libraries and import modules for video processing tasks.

```python
!pip install torch torchvision opencv-python &> /dev/null
from IPython.display import Video

import os
import cv2
import gc
import math
import openai
import textwrap
import torch
import torchvision
```

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

### Install llmcompressor and import modules

Source: https://docs.parasail.io/parasail-docs/products/overview-1/fp8-quantization

Install the necessary library and import the required functions for model quantization.

```bash
# pip install llmcompressor
```

```python
from llmcompressor import oneshot
from llmcompressor.modifiers import
```

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

### Install Dependencies and Convert PDF to Images

Source: https://docs.parasail.io/parasail-docs/guides/multi-modal

Installs necessary system utilities and Python packages, then demonstrates how to iterate through PDF pages to convert them into images.

```python
!apt-get install -qq -y poppler-utils > /dev/null
!pip install pdf2image > /dev/null
import pdf2image

!wget https://arxiv.org/pdf/1706.03762 -O transformer.pdf > /dev/null

# Use multiple images as input
messages = []
for i, page in enumerate(pdf2image.convert_from_path("transformer.pdf",
```

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

### Install and run batch processing script

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

Commands to install the required library and execute a batch processing script with an API key.

```sh
pip3 install openai-batch
PARASAIL_API_KEY=<YOUR-API-KEY> python3 test_omnigen_batch.py
```

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

### Install Dependencies and Import Libraries

Source: https://docs.parasail.io/parasail-docs/guides/multi-modal

Installs necessary packages for computer vision and video processing, followed by standard library imports.

```python
!pip install torch torchvision opencv-python > /dev/null
from IPython.display import Video

import os
import cv2
import gc
import
```

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

### Install Batch Helper Library

Source: https://docs.parasail.io/parasail-docs/api-reference/batch-api

Install the required Python package to facilitate batch processing.

```bash
pip install openai-batch
```

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

### Install OpenAI SDK

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

Install the required OpenAI Python library via pip.

```bash
pip install openai
```

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

### Install dependencies and process PDF for VQA

Source: https://docs.parasail.io/parasail-docs/guides/multi-modal

Install necessary system and Python packages, then convert PDF pages to images for multi-modal input.

```python
!apt-get install -qq -y poppler-utils &> /dev/null
!pip install pdf2image &> /dev/null
import pdf2image

!wget https://arxiv.org/pdf/1706.03762 -O transformer.pdf &> /dev/null

# Use multiple images as input
messages = []
for i, page in enumerate(pdf2image.convert_from_path("transformer.pdf", last_page=4)):
    page.save(f"page_{i}.jpg", 'JPEG')
    messages.append({"type": "image_url", "image_url": { "url": data_url(f"page_{i}.jpg")}})

messages = [{
    "role": "user",
    "content": messages + [{
        "type": "text",
        "text": "How many of encoder layers were used and where in the paper it's mentioned?"
    }]
}]

chat_completion = oai_client.chat.completions.create(
    model="parasail-qwen25-vl-72b-instruct", messages=messages)

display(Image("page_0.jpg", width=500))
print(chat_completion.choices[0].message.content)
```

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

### Batch Processing Example

Source: https://docs.parasail.io/parasail-docs/products/quickstart

Demonstrates the syntax for defining a batch operation.

```text
Batch as batch:
```

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

### Configure Model and Print Output

Source: https://docs.parasail.io/parasail-docs/operate-in-production/retries-and-idempotency

Example of setting a model configuration and printing the result.

```python
model = "<your-model>",
)
print(
```

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

### Configure Model and Batch

Source: https://docs.parasail.io/parasail-docs/products/quickstart

Examples for setting the model identifier and initializing a batch request.

```python
model = "NousResearch/DeepHermes-3-Mistral-24B-Preview",
```

```python
messages = [{
```

```python
batch.add_to_batch(
```

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

### Models API Request Examples

Source: https://docs.parasail.io/parasail-docs/api-reference/models-endpoint

Examples for interacting with the models endpoint using cURL and the Python OpenAI SDK.

```bash
curl https://api.parasail.io/v1/models \
  -H "Authorization: Bearer $PARASAIL_API_KEY"
```

```python
from openai import OpenAI

client = OpenAI(
    base_url="https://api.parasail.io/v1",
    api_key="YOUR_PARASAIL_API_KEY"
)

models = client.models.list()
print(models)
```

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

### API Key Configuration

Source: https://docs.parasail.io/parasail-docs/guides/multi-modal

Example of setting the API key for authentication.

```text
api_key = API_KEY
```

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

### Importing Required Libraries

Source: https://docs.parasail.io/parasail-docs/guides/structured-output

Initial setup for using the OpenAI client with Parasail.

```python
import json
import os
from openai import
```

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

### Importing Parasail Client

Source: https://docs.parasail.io/parasail-docs/guides/tool-function-calling

Initial setup for using the Parasail Python client.

```python
import 
```

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

### Deployment configuration JSON

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

Example JSON structure for defining a deployment configuration.

```json
{
 "id": 12345,
 "deploymentName": "myqwen",
 "modelName":
```

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

### Object Definition Example

Source: https://docs.parasail.io/parasail-docs/products/quickstart

Shows how to define a list of objects for processing.

```text
objects = ["cat", "robot", 
```

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

### Example JSON Output

Source: https://docs.parasail.io/parasail-docs/guides/structured-output

Sample output format for the weather query.

```json
{
  "location": "Manhattan Beach",
  "date": "2025-06-03"
}
```

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

### Configure Model and Prompt Parameters

Source: https://docs.parasail.io/parasail-docs/products/overview-2

Examples showing how to define the model name, input prompt, and size configuration.

```text
model = "Shitao/OmniGen-v1",
```

```text
prompt = "A man in a black shirt is reading a book. The man is the right man in <img><|image_1|></img>",
```

```text
size =
```

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

### Guided Constraints Examples

Source: https://docs.parasail.io/parasail-docs/guides/structured-output

Examples of using guided_regex for pattern matching and guided_choice for restricted selection.

```python
# Constrain to an ISO date
extra_body = {"guided_regex": r"\d{4}-\d{2}-\d{2}"}

# Constrain to one of a fixed set of values
extra_body = {"guided_choice": ["red", "green", "blue"]}
```

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

### Retrieve and Select Hardware Configuration

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

This example demonstrates how to fetch supported hardware configurations, identify a specific one, and update the selected flag to true.

```javascript
Here is example code to retrieve supported hardware, look for a desired configuration, and flip the selected flag to true when it is found:
```

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

### Initialize Device Configuration

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

Example of calling the device configuration selection function with specific model parameters.

```python
devices = get_and_select_device_config(
 client,
 model_name="Qwen/Qwen2.5-72B-Instruct",
```

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

### Billing API JSON Response Structure

Source: https://docs.parasail.io/parasail-docs/api-reference/billing-api

Example of the JSON response format returned by the billing API, showing total usage and start timestamps.

```json
{
  "total": 52.89,
  "start_timestamp": "2026-04-01T00:00:00+00:00"
}
```

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

### Initialize Client

Source: https://docs.parasail.io/parasail-docs/products/overview/model-specific-notes

Basic client initialization syntax.

```text
client =
```

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

### GET /api/v1/dedicated/performance

Source: https://docs.parasail.io/parasail-docs/security-and-account-management/account-api-keys

Get model and hardware performance estimates.

```APIDOC
## GET /api/v1/dedicated/performance

### Description
Get model and hardware performance estimates.

### Method
GET

### Endpoint
/api/v1/dedicated/performance
```

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

### Get Deployment

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

Sends a GET request to retrieve details for a specific deployment by its ID.

```python
def get_deployment(parasail_client: httpx.Client, id: int) -> dict:
 response = parasail_client.get(f"/dedicated/deployments/{id}")
 assert (
 200 <= response.status_code < 300
 ), f"Failed to get deployment: {response.status_code} {response.text}"
 return
```

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

### Batch Processing and Model Configuration

Source: https://docs.parasail.io/parasail-docs/products/overview-2

Examples showing how to add items to a batch and configure a specific model.

```python
batch.add_to_batch()
```

```python
 model = "Shitao/OmniGen-v1",
```

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

### Install OpenAI dependency

Source: https://docs.parasail.io/parasail-docs/guides/multi-modal

Install the required OpenAI Python package to enable multi-modal functionality.

```bash
!pip install openai > /dev/null
```

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

### Batch API Usage

Source: https://docs.parasail.io/parasail-docs/api-reference/batch-api

Demonstrates how to install the library, initialize a batch, add requests, and submit them with metadata.

```APIDOC
## Installation

```bash
pip install openai-batch
```

## Usage

### Description
Use the `Batch` class to queue multiple requests and submit them for processing. The `add_to_batch` method is used to define the model and message content for each request, and `submit` is used to finalize the batch with optional metadata.

### Example
```python
from openai_batch import Batch

with Batch() as batch:
    for i in range(100):
        batch.add_to_batch(
            model="NousResearch/DeepHermes-3-Mistral-24B-Preview",
            messages=[{"role": "user", "content": f"Prompt #{i}"}]
        )

    batch.submit(metadata={"PARASAIL_PRIORITY": ...})
```
```

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

### Initialize Clients and Configure Devices

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

Initializes the Parasail and OpenAI clients and selects a device configuration for a specific model.

```python
 parasail_client = httpx.Client(
 base_url=control_base_url,
 headers={
 "Authorization": "Bearer " + api_key,
 "Content-Type": "application/json",
 },
 )
 openai_client = openai.Client(api_key=api_key, base_url=chat_base_url)
 check_compatibility(parasail_client, model_name)

 devices = get_and_select_device_config(
 parasail_client,
 model_name=model_name,
 desired_device="H100SXM",
 
```

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

### Perform GET Request with Read-Only Key

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

Read-only keys are permitted to perform GET requests on specific management endpoints.

```http
GET /api/v1/dedicated/deployments HTTP/1.1
Host: api.parasail.io
Authorization: Bearer <read-only-api-key>
```

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

### Initialize OpenAI Client for Model Usage

Source: https://docs.parasail.io/parasail-docs/products/overview/model-specific-notes

Basic setup for importing necessary libraries and initializing the OpenAI client for model interaction.

```python
import json
import os

from openai import OpenAI

MODEL =
```

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

### Configure GPT-OSS reasoning parameters

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

Examples showing how to set thinking_budget and reasoning_effort using the OpenAI Python client and cURL.

```python
from openai import OpenAI
import os

client = OpenAI(
    base_url="https://api.parasail.io/v1",
    api_key=os.environ["PARASAIL_API_KEY"],
)

resp = client.chat.completions.create(
    model="parasail-gpt-oss-20b-fast",
    messages=[{"role": "user", "content": "Explain why the sky is blue."}],
    extra_body={
        "custom_params": {"thinking_budget": 40},
        "chat_template_kwargs": {"reasoning_effort": "high"},
    },
)

print(resp.choices[0].message.content)
```

```bash
curl https://api.parasail.io/v1/chat/completions \
  -H "Authorization: Bearer $PARASAIL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "parasail-gpt-oss-20b-fast",
    "messages": [{"role": "user", "content": "Explain why the sky is blue."}],
    "custom_params": {"thinking_budget": 40},
    "chat_template_kwargs": {"reasoning_effort": "high"}
  }'
```

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

### GET /api/v1/openrouter/models

Source: https://docs.parasail.io/parasail-docs/security-and-account-management/account-api-keys

List models available through OpenRouter.

```APIDOC
## GET /api/v1/openrouter/models

### Description
List models available through OpenRouter.

### Method
GET

### Endpoint
/api/v1/openrouter/models
```

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

### Initialize OpenAI Client

Source: https://docs.parasail.io/parasail-docs/guides/chat-completions

Instantiate the OpenAI client for use in your application.

```python
client = OpenAI()
```

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

### GET /v1/billing/invoices

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

List all billing invoices for the organization.

```APIDOC
## GET /v1/billing/invoices

### Description
List invoices.

### Method
GET

### Endpoint
/v1/billing/invoices
```

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

### Initiate Tool-Enabled Request and Agentic Loop

Source: https://docs.parasail.io/parasail-docs/guides/tool-function-calling

Demonstrates how to create an initial request with tool support and process subsequent function calls in a loop.

```python
 return json.dumps({"error": "unknown function"})

# Initial request (store must be false on the Parasail gateway)
response = client.responses.create(
 model="parasail-kimi-k25-elicit",
 store=False,
 tools=tools,
 input=[{"role": "user", "content": "What's the weather in Tokyo?"}],
)

# Agentic loop—continue until the model stops calling tools
while response.output:
 tool_calls = [item for item in response.output if item.type == "function_call"]
 if not tool_calls:
```

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

### GET /api/v1/dedicated/devices

Source: https://docs.parasail.io/parasail-docs/security-and-account-management/account-api-keys

List supported GPU and hardware types.

```APIDOC
## GET /api/v1/dedicated/devices

### Description
List supported GPU and hardware types.

### Method
GET

### Endpoint
/api/v1/dedicated/devices
```

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

### GET /api/v1/dedicated/deployments

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

List all dedicated deployments associated with the organization.

```APIDOC
## GET /api/v1/dedicated/deployments

### Description
List your dedicated deployments.

### Method
GET

### Endpoint
/api/v1/dedicated/deployments
```

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

### Manage and Test Deployment Lifecycle

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

Demonstrates how to verify deployment status, run tests, and toggle deployment states between online and offline.

```python
 wait_for_status(parasail_client, deployment["id"], status="ONLINE")

 print("\nTesting deployment with 1 replica...")
 test_deployment(parasail_client, openai_client, deployment["id"])
 print("Initial deployment tests passed successfully")

 pause_deployment(parasail_client, deployment["id"])
 wait_for_status(parasail_client, deployment["id"], status="OFFLINE")
 resume_deployment(parasail_client, deployment["id"])
 wait_for_status(parasail_client, deployment["id"], status="ONLINE")

 print("\nTesting deployment with 1 replica...")
 test_deployment(parasail_client, openai_client, deployment["id"])
 print("Initial deployment tests passed
```

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

### GET /dedicated/deployments/{id}

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

Retrieves details of a specific deployment.

```APIDOC
## GET /dedicated/deployments/{id}

### Description
Fetches the current status and configuration of a deployment.

### Method
GET

### Endpoint
/dedicated/deployments/{id}

### Parameters
#### Path Parameters
- **id** (integer) - Required - The deployment ID.
```

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

### GET /dedicated/devices

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

Retrieves available device configurations for a model.

```APIDOC
## GET /dedicated/devices

### Description
Retrieves a list of available device configurations for a given model and engine.

### Method
GET

### Endpoint
/dedicated/devices

### Parameters
#### Query Parameters
- **modelName** (string) - Required - The name of the model.
- **engine** (string) - Required - The engine type.
- **modelAccessKey** (string) - Optional - Access key for privately hosted models.
```

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

### GET /dedicated/support

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

Checks if a specific model is supported for deployment.

```APIDOC
## GET /dedicated/support

### Description
Checks if a model is supported for deployment on the platform.

### Method
GET

### Endpoint
/dedicated/support

### Parameters
#### Query Parameters
- **modelName** (string) - Required - The name of the model to check.
- **engine** (string) - Required - The engine type (e.g., VLLM).
- **modelAccessKey** (string) - Optional - Access key for privately hosted models.
```

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

### GET /v1/models

Source: https://docs.parasail.io/parasail-docs/api-reference/chat-completions

Lists the available models that can be used with the API.

```APIDOC
## GET /v1/models

### Description
Retrieves a list of models available for use.

### Method
GET

### Endpoint
https://api.parasail.io/v1/models
```

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

### Batch Job Output Example

Source: https://docs.parasail.io/parasail-docs/guides/rag

Sample console output indicating successful batch job completion and database creation.

```text
Batch job batch-aqjf7etmdt completed successfully

Vector DB is created
Number of embedding vectors in DB:  232
Number of embeddings tokens: 217826
```

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

### Create a model deployment with Python

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

Uses the client to send a POST request to the /dedicated/deployments endpoint. Requires a pre-defined devices configuration.

```python
# Example: Creating a deployment
deployment_config = {
    "deploymentName": "myqwen",
    "modelName": "Qwen/Qwen2.5-72B-Instruct",
    "engine": "VLLM",  # Optional (defaults to VLLM)
    "deviceConfigs": devices,
    "replicas": 1,
    # Optional parameters:
    # "scaleDownPolicy": "TIMER",
    # "scaleDownThreshold": 3600000,
}

response = client.post("/dedicated/deployments", json=deployment_config)

if response.status_code == 200:
    deployment_data = response.json()
    deployment_id = deployment_data["id"]
    print(f"Created deployment with ID: {deployment_id}")
else:
    print(f"Failed to create deployment: {response.status_code}, {response.text}")
```

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

### Python Loop Example

Source: https://docs.parasail.io/parasail-docs/products/quickstart

A simple Python loop structure.

```python
for i in range(100):
```

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

### GET /api/v1/dedicated/support

Source: https://docs.parasail.io/parasail-docs/security-and-account-management/account-api-keys

Check whether a model supports dedicated hardware.

```APIDOC
## GET /api/v1/dedicated/support

### Description
Check whether a model supports dedicated hardware.

### Method
GET

### Endpoint
/api/v1/dedicated/support
```

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

### Create a deployment

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

Sends a POST request to initialize a new model deployment with specific configuration parameters.

```python
deployment_config = {
    "deploymentName": "myqwen",
    "modelName": "Qwen/Qwen2.5-72B-Instruct",
    "engine": "VLLM",  # Optional (defaults to VLLM)
    "deviceConfigs": devices,
    "replicas": 1,
    # Optional parameters:
    # "scaleDownPolicy": "TIMER",
    # "scaleDownThreshold": 3600000,
}

response = client.post("/dedicated/deployments", json=deployment_config)

if response.status_code == 200:
    deployment_data = response.json()
    deployment_id = deployment_data["id"]
    print(f"Created deployment with ID: {deployment_id}")
else:
    print(f"Failed to create deployment: {response.status_code}, {response.text}")
```

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

### GET /v1/billing/invoices/current

Source: https://docs.parasail.io/parasail-docs/api-reference/billing-api

Retrieves the current billing invoice for the authenticated account.

```APIDOC
## GET /v1/billing/invoices/current

### Description
Retrieves the current billing invoice for the authenticated account.

### Method
GET

### Endpoint
/v1/billing/invoices/current
```

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

### GET /invoices/{id}

Source: https://docs.parasail.io/parasail-docs/api-reference/billing-api

Retrieves a specific invoice by its unique identifier.

```APIDOC
## GET /invoices/{id}

### Description
Returns a specific invoice by its unique identifier.

### Method
GET

### Endpoint
/invoices/{id}

### Parameters
#### Path Parameters
- **id** (string) - Required - The unique identifier of the invoice.
```

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

### Environment and model configuration setup

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

Initializes environment variables and defines Pydantic models for device configuration and model creation.

```python
import httpx
import openai
from pydantic import BaseModel
from typing import Optional
import os
import time
from dotenv import load_dotenv

# Load environment variables
env_path = os.path.expanduser("~/dev/parasail/.env")
if os.path.exists(env_path):
    load_dotenv(env_path)
api_key = os.environ["PARASAIL_API_KEY"]
control_base_url = "https://api.parasail.io/api/v1"
chat_base_url = "https://api.parasail.io/v1"


class deviceConfig(BaseModel):
    device: str
    count: int
    displayName: Optional[str] = None
    cost: Optional[float] = None
    estimatedSingleUserTps: Optional[float] = None
    estimatedSystemTps: Optional[float] = None
    recommended: Optional[bool] = None
    limitedContext: Optional[bool] = None
    available: Optional[bool] = None
    selected: bool


class CreateModelStruct(BaseModel):
```

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

### GET /v1/billing/invoices/breakdowns

Source: https://docs.parasail.io/parasail-docs/api-reference/billing-api

Retrieves usage breakdowns for a specified time range.

```APIDOC
## GET /v1/billing/invoices/breakdowns

### Description
Retrieves usage breakdowns for a specified time range.

### Method
GET

### Endpoint
/v1/billing/invoices/breakdowns

### Parameters
#### Query Parameters
- **starting_on** (string) - Required - Start timestamp for the range.
- **ending_before** (string) - Required - End timestamp for the range.
- **window_size** (string) - Optional - The granularity of the breakdown (e.g., day).
- **limit** (integer) - Optional - Maximum number of results to return.
- **next_page** (string) - Optional - Token for pagination.

### Request Example
curl -H "Authorization: Bearer psk-<accessKey>-<secretKey>" \
  "https://api.parasail.io/v1/billing/invoices/breakdowns?starting_on=2026-05-01T00:00:00Z&ending_before=2026-06-01T00:00:00Z&window_size=day&limit=35&next_page=..."
```

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

### GET /api/v1/dedicated/deployments/{id}

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

Fetch details for a specific dedicated deployment.

```APIDOC
## GET /api/v1/dedicated/deployments/{id}

### Description
Fetch one dedicated deployment.

### Method
GET

### Endpoint
/api/v1/dedicated/deployments/{id}

### Parameters
#### Path Parameters
- **id** (string) - Required - The unique identifier of the deployment.
```

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

### GET /models

Source: https://docs.parasail.io/parasail-docs/api-reference/models-endpoint

Retrieves a list of available models that can be used for inference.

```APIDOC
## GET /models

### Description
Retrieves a list of available models, including serverless models, dedicated deployments, and HuggingFace models.

### Method
GET

### Endpoint
/models

### Response
#### Success Response (200)
- **object** (string) - The type of the object, which is 'list'.
- **data** (array) - A list of model objects.

#### Response Example
{
  "object": "list",
  "data": [
    {
      "id": "parasail-deepseek-r1",
      "object": "model",
      "created": 1727900000,
      "owned_by": "parasail"
    }
  ]
}
```

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

### Client Setup

Source: https://docs.parasail.io/parasail-docs/api-reference/batch-api

How to configure the OpenAI Python client to communicate with the Parasail Batch API.

```APIDOC
## Client Setup

### Description
To use the Parasail Batch API, configure the OpenAI client with the Parasail base URL and your Parasail API key.

### Code Example
```python
from openai import OpenAI

client = OpenAI(
    base_url="https://api.parasail.io/v1",
    api_key="<PARASAIL_API_KEY>",
)
```
```

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

### Create JSONL Batch Input File in Python

Source: https://docs.parasail.io/parasail-docs/products/quickstart/file-format

Example demonstrating how to write a list of prompts into a JSONL file format suitable for batch processing.

```python
import json

prompts = ["What is the capital of New York?", "Write two very funny dad jokes.", "Say pong."]

with open("example_batch_input.jsonl", "w") as file:
 for i, prompt in enumerate(prompts):
  file.write(
 json.dumps(
```

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

### Batch Processing Syntax

Source: https://docs.parasail.io/parasail-docs/quickstart/batch

Example of defining a batch operation in Python.

```python
with Batch() as batch:
    objects = ["cat"]
```

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

### Update Deployment Configuration

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

Demonstrates how to fetch an existing deployment, modify its replica count, and push the update back to the server.

```python
# 1. Fetch the current deployment
deployment_id = 12345
response = client.get(f"/dedicated/deployments/{deployment_id}")
current_deployment = response.json()

# 2. Modify the "replicas" field (for example, from 1 to 2)
current_deployment["replicas"] = 2

# 3. Send the updated deployment object
update_response = client.put(
 f"/dedicated/deployments/{deployment_id}",
```

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

### Define Tool Function

Source: https://docs.parasail.io/parasail-docs/guides/tool-function-calling

Example of defining a tool function in Python.

```python
def get_weather(location, 
```

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

### Initialize Parasail API Client

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

Basic setup for the API client using the httpx library and your API key.

```python
import httpx

api_key = "<YOUR_PARASAIL_API_KEY>"
```

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

### Execute Video Chat Completion

Source: https://docs.parasail.io/parasail-docs/llms-full.txt

Example workflow for downloading a video, processing it, and sending it to a multimodal model for description.

```python
!wget https://upload.wikimedia.org/wikipedia/commons/2/2f/Making_snowman_in_K%C3%B5rvemaa%2C_Estonia_%28January_2022%29.webm -O snowman.webm &> /dev/null

print(f"Processing snowman.webm...")
Video("snowman.webm", embed=True)
video_url, frames = process_video("snowman.webm")

# Generate chat completions for each model/prompt pair.
chat_response = oai_client.chat.completions.create(
    model="parasail-qwen25-vl-72b-instruct",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "video_url", "video_url": {"url": video_url}},
                {"type": "text", "text": "Describe the video in detail"},
            ],
        }
    ],
    temperature=1,
    top_p=0.001,
    max_tokens=4096,
    extra_body={"repetition_penalty": 1.05},
)
print("---------------------------")
print(textwrap.fill(f"output={chat_response.choices[0].message.content}"))
print("=============================")
```

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

### Initialize OpenAI Client for Chat Completions

Source: https://docs.parasail.io/parasail-docs/guides/chat-completions

Basic import statements required to begin using the OpenAI client with Parasail.

```python
import os
from openai import OpenAI
```

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

### Import and File Handling in Python

Source: https://docs.parasail.io/parasail-docs/products/overview-2

Demonstrates the basic import of base64 and the use of the with open statement for file operations.

```python
import base64

with open
```

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

### GET /dedicated/support

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

Queries whether a specific model is supported by the Parasail platform.

```APIDOC
## GET /dedicated/support

### Description
Checks if a model is supported for dedicated deployment. The model can be identified by a Hugging Face ID or URL.

### Method
GET

### Endpoint
/dedicated/support

### Parameters
#### Query Parameters
- **modelName** (string) - Required - The Hugging Face ID or URL of the model.
- **engine** (string) - Optional - The engine to use (e.g., VLLM).
```

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

### GET /api/v1/dedicated/deployments/{id}/models

Source: https://docs.parasail.io/parasail-docs/security-and-account-management/account-api-keys

List models on one dedicated deployment.

```APIDOC
## GET /api/v1/dedicated/deployments/{id}/models

### Description
List models on one dedicated deployment.

### Method
GET

### Endpoint
/api/v1/dedicated/deployments/{id}/models

### Parameters
#### Path Parameters
- **id** (string) - Required - The ID of the deployment.
```

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

### Importing OpenAI Client

Source: https://docs.parasail.io/parasail-docs/api-reference/embeddings

Initial setup for using the OpenAI client library within a Python environment.

```python
import os
from openai 
```

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

### Retrieve Billing Record

Source: https://docs.parasail.io/parasail-docs/api-reference/billing-api

Standard GET request to fetch billing information.

```http
GET
```

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

### Retrieve and Select Device Configuration

Source: https://docs.parasail.io/parasail-docs/products/overview-1/management-api

Fetches available device configurations and marks the desired one as selected.

```python
def get_and_select_device_config(
    parasail_client: httpx.Client,
    model_name: str,
    desired_device: str,
    desired_count: int,
) -> list[dict]:
    params = {"engine": "VLLM", "modelName": model_name, "modelAccessKey": ""}

    response = parasail_client.get("/dedicated/devices", params=params)
    if response.status_code != 200:
        raise RuntimeError(f"Request failed: {response.status_code} - {response.text}")

    devices = response.json()
    matched = False

    for d in devices:
        d["selected"] = (
            d.get("device") == desired_device and d.get("count") == desired_count
        )
        matched = matched or d["selected"]

    if not matched:
        raise ValueError(
            f"No matching device config for {desired_device} x{desired_count}"
        )

    return devices
```