### Get Available Models (Python)

Source: https://vsegpt.ru/Docs/API/Code

Example of fetching a list of all available models supported by the Vsegpt API. This helps in choosing the right model for your task.

```python
from vsegpt import Vsegpt

client = Vsegpt("YOUR_API_KEY")

models = client.models.list()
for model in models.data:
    print(model.id)
```

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

### Third-Party Libraries and Examples (C# async)

Source: https://vsegpt.ru/Docs/API/Code

An asynchronous example in C# for interacting with Vsegpt, suitable for .NET applications. Demonstrates non-blocking API calls.

```csharp
using System;
using System.Net.Http;
using System.Text;
using System.Text.Json;
using System.Threading.Tasks;

public class VsegptClient
{
    private readonly HttpClient _httpClient;
    private readonly string _apiKey;

    public VsegptClient(string apiKey)
    {
        _httpClient = new HttpClient();
        _apiKey = apiKey;
        _httpClient.DefaultRequestHeaders.Authorization = new System.Net.Http.Headers.AuthenticationHeaderValue("Bearer", _apiKey);
    }

    public async Task<string> GetChatCompletionAsync(string prompt)
    {
        var requestData = new
        {
            model = "gpt-4o-mini",
            messages = new[]
            {
                new { role = "system", content = "You are a helpful assistant." },
                new { role = "user", content = prompt }
            }
        };

        var jsonContent = JsonSerializer.Serialize(requestData);
        var httpContent = new StringContent(jsonContent, Encoding.UTF8, "application/json");

        var response = await _httpClient.PostAsync("https://api.vsegpt.ai/v1/chat/completions", httpContent);
        response.EnsureSuccessStatusCode();

        var responseString = await response.Content.ReadAsStringAsync();
        var responseObject = JsonSerializer.Deserialize<dynamic>(responseString);
        return responseObject.choices[0].message.content;
    }

    public static async Task Main(string[] args)
    {
        var client = new VsegptClient("YOUR_API_KEY");
        var completion = await client.GetChatCompletionAsync("Hello!");
        Console.WriteLine(completion);
    }
}
```

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

### Third-Party Libraries and Examples (NodeJS)

Source: https://vsegpt.ru/Docs/API/Code

Provides a NodeJS example for interacting with third-party libraries or services that might integrate with Vsegpt. Useful for web development.

```javascript
const axios = require('axios');

const apiKey = 'YOUR_API_KEY';
const apiUrl = 'https://api.vsegpt.ai/v1/chat/completions';

axios.post(apiUrl, {
    model: 'gpt-4o-mini',
    messages: [
        {role: 'system', content: 'You are a helpful assistant.'},
        {role: 'user', content: 'Hello!'}
    ]
}, {
    headers: {
        'Authorization': `Bearer ${apiKey}`,
        'Content-Type': 'application/json'
    }
})
.then(response => {
    console.log(response.data.choices[0].message.content);
})
.catch(error => {
    console.error('Error:', error);
});
```

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

### Basic Example - gpt-4o-mini Call (Google Colab)

Source: https://vsegpt.ru/Docs/API/Code

A basic example in Google Colab demonstrating how to call the gpt-4o-mini model. This is a starting point for users of the platform.

```python
# Install the Vsegpt library
!pip install vsegpt

from vsegpt import Vsegpt

client = Vsegpt("YOUR_API_KEY")

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello!"}
    ]
)

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

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

### Get Account Balance (Python)

Source: https://vsegpt.ru/Docs/API/Code

Example of retrieving your current account balance using the Vsegpt API in Python. This is essential for monitoring usage and costs.

```python
from vsegpt import Vsegpt

client = Vsegpt("YOUR_API_KEY")

balance = client.billing.get_balance()
print(f"Your current balance is: {balance.amount}")
```

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

### Third-Party Libraries and Examples (TypeScript with OpenAI library)

Source: https://vsegpt.ru/Docs/API/Code

Example using the official OpenAI library in TypeScript to interact with Vsegpt. This leverages existing SDKs for convenience.

```typescript
import OpenAI from 'openai';

const openai = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://api.vsegpt.ai/v1',
});

async function main() {
  const completion = await openai.chat.completions.create({
    model: "gpt-4o-mini",
    messages: [
      {"role": "system", "content": "You are a helpful assistant."}, 
      {"role": "user", "content": "Hello!"}
    ]
  });

  console.log(completion.choices[0].message.content);
}

main();
```

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

### Basic API Call (Python)

Source: https://vsegpt.ru/Docs/API/Code

A fundamental example demonstrating how to make a basic API call using Python. This snippet is useful for getting started with text generation.

```python
from vsegpt import Vsegpt

client = Vsegpt("YOUR_API_KEY")

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello!"}
    ]
)

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

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

### FLUX.2 Pro Image Generation Examples

Source: https://vsegpt.ru/Docs/Models/Txt2Img

Examples demonstrating text-to-image generation with FLUX.2 Pro, including HEX color usage and image referencing.

```string
"a wall painted in color #2ECC71"
```

```string
"gradient from hex #FF6B6B to hex #4ECDC4"
```

```string
"the car in color #1A1A1A with accents in #FFD700"
```

```string
"@image1 wearing the outfit from @image2"
```

```string
"combine the style of @image1 with the composition of @image2"
```

```string
"the person from @image1 in the setting from @image3"
```

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

### Get Reasoning in Reasoning Models (Python)

Source: https://vsegpt.ru/Docs/API/Code

Example of retrieving reasoning outputs from models like o1 and R1 using the Vsegpt API. This is useful for understanding the model's thought process.

```python
from vsegpt import Vsegpt

client = Vsegpt("YOUR_API_KEY")

response = client.chat.completions.create(
    model="o1", # or "R1"
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Solve this math problem: 2 + 2"}
    ],
    reasoning=True
)

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

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

### Flux 2 [klein] 4B Advanced Prompting Examples

Source: https://vsegpt.ru/Docs/Models/Txt2Img

Examples demonstrating advanced prompting techniques for the Flux 2 [klein] 4B model, including HEX color control, image referencing with '@', and structured JSON prompts for detailed scene control.

```text
"a wall painted in color #2ECC71"
"gradient from hex #FF6B6B to hex #4ECDC4"
"the car in color #1A1A1A with accents in #FFD700"
```

```text
"@image1 wearing the outfit from @image2"
"combine the style of @image1 with the composition of @image2"
"the person from @image1 in the setting from @image3"
```

```json
{
"scene": "Overall setting description",
"subjects": [
{
"type": "Subject category",
"description": "Physical attributes and details",
"pose": "Action or stance",
"position": "foreground/midground/background"
}
],
"style": "Artistic rendering approach",
"color_palette": ["color1", "color2", "color3"],
"lighting": "Lighting conditions and direction",
"mood": "Emotional atmosphere",
"composition": "rule of thirds/centered/dynamic diagonal",
"camera": {
"angle": "eye level/low angle/high angle",
"distance": "close-up/medium shot/wide shot",
"lens": "35mm/50mm/85mm"
}
}
```

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

### Basic API Call (CURL)

Source: https://vsegpt.ru/Docs/API/Code

A fundamental example demonstrating how to make a basic API call using CURL. This snippet is useful for getting started with text generation.

```bash
curl -X POST "https://api.vsegpt.ai/v1/chat/completions" \
     -H "Authorization: Bearer YOUR_API_KEY" \
     -H "Content-Type: application/json" \
     -d '{ 
         "model": "gpt-4o-mini", 
         "messages": [
             {"role": "system", "content": "You are a helpful assistant."}, 
             {"role": "user", "content": "Hello!"}
         ]
     }'
```

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

### Python Tool Call Example (Recommended)

Source: https://vsegpt.ru/Docs/API/AddFeatures

This example shows how to use the recommended tool calls feature, which is compatible with OpenAI and other models supporting the OpenAI format. It defines a tool with a function and passes it to the ChatCompletion API.

```python
#
    prompt = "What is the weather like in Boston?"

    messages = []
    #messages.append({"role": "system", "content": system_text})
    messages.append({"role": "user", "content": prompt})

    response = openai.ChatCompletion.create(

        model="openai/gpt-3.5-turbo",
        messages =messages,
        temperature=0.1,
        n=1,
        max_tokens = 1500,
        stream=False,
        tool_choice="auto",
        tools = [{
            "type": "function",
            "function": {
                "name": "get_current_weather",
                "description": "Get the current weather in a given location",
                "parameters": {
                    "type": "object",
                    "properties": {
                        "location": {
                            "type": "string",
                            "description": "The city and state, e.g. San Francisco, CA",
                        },
                        "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]},
                    },
                    "required": ["location"],
                },
            },
        }
        ]
    )

```

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

### Third-Party Libraries and Examples (TypeScript)

Source: https://vsegpt.ru/Docs/API/Code

A TypeScript example for interacting with the Vsegpt API. Suitable for modern JavaScript development, including Node.js and browser applications.

```typescript
import axios from 'axios';

const apiKey = 'YOUR_API_KEY';
const apiUrl = 'https://api.vsegpt.ai/v1/chat/completions';

async function getCompletion(prompt: string): Promise<string> {
    try {
        const response = await axios.post(apiUrl, {
            model: 'gpt-4o-mini',
            messages: [
                {role: 'system', content: 'You are a helpful assistant.'},
                {role: 'user', content: prompt}
            ]
        }, {
            headers: {
                'Authorization': `Bearer ${apiKey}`,
                'Content-Type': 'application/json'
            }
        });
        return response.data.choices[0].message.content;
    } catch (error) {
        console.error('Error:', error);
        throw error;
    }
}

getCompletion('Hello!').then(content => console.log(content));
```

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

### Generate Video from Image (Image-to-Video) (Python)

Source: https://vsegpt.ru/Docs/API/Code

Example of generating video content from an input image using the Vsegpt API. This allows for animating static images.

```python
from vsegpt import Vsegpt

client = Vsegpt("YOUR_API_KEY")

response = client.videos.generate_from_image(
    image=open("path/to/your/image.png", "rb"),
    prompt="Animate this image to show a gentle breeze blowing through the trees.",
    model="image-to-video",
    size="720p",
    fps=24
)

video_url = response.data[0].url
print(video_url)
```

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

### Generate Music (Text-to-Audio) (Python)

Source: https://vsegpt.ru/Docs/API/Code

Example of generating music from text prompts using the Vsegpt API. This feature allows for creating custom audio tracks.

```python
from vsegpt import Vsegpt

client = Vsegpt("YOUR_API_KEY")

response = client.audio.generate_music(
    prompt="Uplifting electronic music with a driving beat",
    model="text-to-audio",
    duration_seconds=30
)

audio_url = response.data[0].url
print(audio_url)
```

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

### Dialogue Support (Python)

Source: https://vsegpt.ru/Docs/API/Code

Example of how to maintain a conversation history for dialogue support using the Vsegpt API in Python.

```python
from vsegpt import Vsegpt

client = Vsegpt("YOUR_API_KEY")

messages = [
    {"role": "system", "content": "You are a helpful assistant."}
]

while True:
    user_input = input("User: ")
    if user_input.lower() == 'exit':
        break
    messages.append({"role": "user", "content": user_input})
    
    response = client.chat.completions.create(
        model="gpt-4o-mini",
        messages=messages
    )
    
    assistant_message = response.choices[0].message.content
    print(f"Assistant: {assistant_message}")
    messages.append({"role": "assistant", "content": assistant_message})
```

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

### Libraries for Other Languages (Go)

Source: https://vsegpt.ru/Docs/API/Code

Information and potential examples for using a Vsegpt library for the Go programming language. Useful for Go developers integrating with the API.

```go
// Example using a hypothetical Go client library
// package main
// 
// import (
// 	"fmt"
// 	"log"
// 	"your_vsegpt_go_library"
// )
// 
// func main() {
// 	client := your_vsegpt_go_library.NewClient("YOUR_API_KEY")
// 	
// 	resp, err := client.Chat.Completions(context.Background(), &your_vsegpt_go_library.ChatCompletionRequest{
// 		Model: "gpt-4o-mini",
// 		Messages: []your_vsegpt_go_library.Message{
// 			{Role: "system", Content: "You are a helpful assistant."},
// 			{Role: "user", Content: "Hello!"},
// 		},
// 	})
// 	if err != nil {
// 		log.Fatalf("Error: %v", err)
// 	}
// 	
// 	fmt.Println(resp.Choices[0].Message.Content)
// }
```

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

### Speech Recognition (Python)

Source: https://vsegpt.ru/Docs/API/Code

Example of using the Vsegpt API for speech recognition (Speech-to-Text). This converts audio input into text.

```python
from vsegpt import Vsegpt

client = Vsegpt("YOUR_API_KEY")

with open("path/to/your/audio.wav", "rb") as audio_file:
    transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=audio_file
    )

print(transcript.text)
```

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

### DeepSeek V3.2 671B Example

Source: https://vsegpt.ru/Docs/ModelsNew

An updated version of DeepSeek V3.1 featuring a new architecture for lower cost and comparable quality. Supports tools in the standard OpenAI format.

```json
{
  "model": "deepseek-v3.2-671b",
  "messages": [
    {"role": "system", "content": "You are a helpful assistant."}, 
    {"role": "user", "content": "Translate 'Hello, world!' to French."} 
  ],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "translate_text",
        "description": "Translates text from one language to another",
        "parameters": {
          "type": "object",
          "properties": {
            "text": {
              "type": "string",
              "description": "The text to translate"
            },
            "target_language": {
              "type": "string",
              "description": "The language to translate to"
            }
          },
          "required": ["text", "target_language"]
        }
      }
    }
  ],
  "tool_choice": "auto"
}
```

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

### OpenAI GPT-5.1 Codex Max Example

Source: https://vsegpt.ru/Docs/ModelsNew

This model is designed for long-term software development tasks with high context. It supports tools and structured outputs in the standard OpenAI format. Maximum output generation size is 128,000 tokens.

```json
{
  "model": "gpt-5.1-codex-max",
  "messages": [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Write a Python function to calculate the factorial of a number."} 
  ],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "calculate_factorial",
        "description": "Calculates the factorial of a non-negative integer",
        "parameters": {
          "type": "object",
          "properties": {
            "number": {
              "type": "integer",
              "description": "The non-negative integer to calculate the factorial for"
            }
          },
          "required": ["number"]
        }
      }
    }
  ],
  "tool_choice": "auto"
}
```

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

### DeepSeek V3.2 Speciale 671B Example

Source: https://vsegpt.ru/Docs/ModelsNew

A special version of DeepSeek with enhanced reasoning quality, known for its performance in math and computer science competitions. Supports tools in the standard OpenAI format.

```json
{
  "model": "deepseek-v3.2-speciale-671b",
  "messages": [
    {"role": "system", "content": "You are a math expert."}, 
    {"role": "user", "content": "Solve the following equation: 2x + 5 = 15"}
  ],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "solve_equation",
        "description": "Solves a linear equation",
        "parameters": {
          "type": "object",
          "properties": {
            "equation": {
              "type": "string",
              "description": "The equation to solve"
            }
          },
          "required": ["equation"]
        }
      }
    }
  ],
  "tool_choice": "auto"
}
```

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

### DeepSeek V3.2 671B (Thinking) Example

Source: https://vsegpt.ru/Docs/ModelsNew

An updated version of DeepSeek V3.1 with a new architecture, offering reduced cost and similar quality. This version is optimized for reasoning and supports tools in the standard OpenAI format.

```json
{
  "model": "deepseek-v3.2-671b-thinking",
  "messages": [
    {"role": "system", "content": "You are a helpful assistant."}, 
    {"role": "user", "content": "Explain the concept of recursion."} 
  ],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "explain_concept",
        "description": "Explains a given concept",
        "parameters": {
          "type": "object",
          "properties": {
            "concept": {
              "type": "string",
              "description": "The concept to explain"
            }
          },
          "required": ["concept"]
        }
      }
    }
  ],
  "tool_choice": "auto"
}
```

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

### Test Structured Outputs with JSON Schema

Source: https://vsegpt.ru/Docs/API/AddFeatures

Demonstrates how to use structured outputs with a JSON schema for a math tutor prompt. This example requires the 'openai' library and specifies a detailed schema for the response.

```python
def test_structured_outputs(model:str) -> dict:
    math_tutor_prompt = '''
        You are a helpful math tutor. You will be provided with a math problem,
        and your goal will be to output a step by step solution, along with a final answer.
        For each step, just provide the output as an equation use the explanation field to detail the reasoning.
    '''

    prompt = "There are equation 'x+3=8'. Find the integer 'x'"

    messages = []

    messages.append({"role": "system", "content": math_tutor_prompt})
    messages.append({"role": "user", "content": prompt})

    response = openai.ChatCompletion.create(
        headers={
            "X-Title": "Test structured outputs"
        },
        model=model,
        messages=messages,
        temperature=0.1,
        n=1,
        max_tokens=500,
        response_format={
            "type": "json_schema",
            "json_schema": {
                "name": "math_reasoning",
                "schema": {
                    "type": "object",
                    "properties": {
                        "steps": {
                            "type": "array",
                            "items": {
                                "type": "object",
                                "properties": {
                                    "explanation": {"type": "string"},
                                    "output": {"type": "string"}
                                },
                                "required": ["explanation", "output"],
                                "additionalProperties": False
                            }
                        },
                        "final_answer": {"type": "integer"}
                    },
                    "required": ["steps", "final_answer"],
                    "additionalProperties": False
                },
                "strict": True
            }
        }

    )
    print("Response:", response)
    response1 = response["choices"][0]["message"] # будет содержать content, в котором нужный JSON со структурой

    # response_text: str = str(response1).strip()
    return response1
```

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

### Python Function Call Example (Deprecated)

Source: https://vsegpt.ru/Docs/API/AddFeatures

This example demonstrates how to use the deprecated function calls feature with OpenAI models. It shows how to define a function and pass it to the ChatCompletion API.

```python
#
    prompt = "What is the weather like in Boston?"

    messages = []
    #messages.append({"role": "system", "content": system_text})
    messages.append({"role": "user", "content": prompt})

    response = openai.ChatCompletion.create(

        model="openai/gpt-3.5-turbo",
        # model="jondurbin/airoboros-l2-70b",
        messages =messages,
        temperature=0.9,
        n=1,
        max_tokens = 1500,
        stream=stream,
        functions=[
            {
                "name": "get_current_weather",
                "description": "Get the current weather in a given location",
                "parameters": {
                    "type": "object",
                    "properties": {
                        "location": {
                            "type": "string",
                            "description": "The city and state, e.g. San Francisco, CA"
                        },
                        "unit": {
                            "type": "string",
                            "enum": ["celsius", "fahrenheit"]
                        }
                    },
                    "required": ["location"]
                }
            }
        ]
    )

```

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

### C# Async API Call

Source: https://vsegpt.ru/Docs/API/Code

This C# example shows how to make an asynchronous API call to VseGPT using HttpClient. It includes setting up the request with messages, model, and other parameters, and handling the response.

```csharp
static async Task Main()
        {

            string apiKey = "sk-or-vv-86*************"; // ваш ключ в VseGPT после регистрации
            string baseApiUrl = "https://api.vsegpt.ru/v1/";
            string prompt = "Напиши последовательно числа от 1 до 10";

            try
            {
                HttpClient client = new HttpClient();
                client.DefaultRequestHeaders.Add("Authorization", $"Bearer {apiKey}");

                List<dynamic> messages = new List<dynamic>();
                messages.Add(new { role = "user", content = prompt });

                var requestData = new
                {
                    model = "openai/gpt-4o-mini",
                    messages = messages,
                    temperature = 0.7,
                    n = 1,
                    max_tokens = Convert.ToInt32(prompt.Length * 1.5),
                    extra_headers = new { X_Title = "My App" } // опционально - передача информации об источнике API-вызова
                };

                var jsonRequest = Newtonsoft.Json.JsonConvert.SerializeObject(requestData);
                var content = new StringContent(jsonRequest, Encoding.UTF8, "application/json");

                var response = await client.PostAsync(baseApiUrl + "chat/completions", content);

                if (response.IsSuccessStatusCode)
                {
                    var jsonResponse = await response.Content.ReadAsStringAsync();
                    dynamic responseData = Newtonsoft.Json.JsonConvert.DeserializeObject(jsonResponse);
                    string responseContent = responseData.choices[0].message.content;
                    Console.WriteLine("Response: " + responseContent);
                }
                else
                {
                    Console.WriteLine("Error: " + response.ReasonPhrase);
                }
            }
            catch (Exception ex)
            {
                Console.WriteLine("Exception: " + ex.Message);
            }
}
```

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

### Working with Large Text - RAG Example (Google Colab)

Source: https://vsegpt.ru/Docs/API/Code

An example in Google Colab demonstrating Retrieval-Augmented Generation (RAG) for searching a knowledge base using embeddings. This is useful for question-answering on custom data.

```python
# Install necessary libraries
!pip install vsegpt langchain-community

from vsegpt import Vsegpt
from langchain_community.embeddings import VsegptEmbeddings
from langchain_community.vectorstores import FAISS
from langchain_core.documents import Document

client = Vsegpt("YOUR_API_KEY")

# Sample knowledge base
knowledge_base = [
    "Vsegpt is a powerful AI model.",
    "It supports various tasks like text generation and embeddings.",
    "RAG stands for Retrieval-Augmented Generation."
]

# Generate embeddings
embeddings = VsegptEmbeddings(client=client, model="text-embedding-ada-002")

# Create a vector store
documents = [Document(page_content=text) for text in knowledge_base]
vector_store = FAISS.from_documents(documents, embeddings)

# Query the knowledge base
query = "What is Vsegpt?"
results = vector_store.similarity_search(query)

# Use the retrieved information to generate an answer (simplified)
context = " ".join([doc.page_content for doc in results])

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[
        {"role": "system", "content": f"Use the following context to answer the question. Context: {context}"},
        {"role": "user", "content": query}
    ]
)

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