Input Formats

Learn about the supported input formats for Rustellar AI models.

Request Structure

All API requests follow the OpenAI-compatible format for easy integration.

Basic Request Format

{
  "model": "helix-v1",
  "messages": [
    {
      "role": "system",
      "content": "You are a helpful assistant."
    },
    {
      "role": "user",
      "content": "Hello!"
    }
  ],
  "temperature": 0.7,
  "max_tokens": 1000
}

Message Roles

System

System messages set the behavior and context for the AI model.

{
  "role": "system",
  "content": "You are a helpful assistant specialized in programming."
}

User

User messages represent the input from the end user.

{
  "role": "user",
  "content": "What is Python?"
}

Assistant

Assistant messages represent previous responses from the AI model. Used for multi-turn conversations.

{
  "role": "assistant",
  "content": "Python is a high-level programming language..."
}

Parameters

Required Parameters

Parameter	Type	Description
model	string	The model to use (e.g., "helix-v1", "iroha")
messages	array	Array of message objects with role and content

Optional Parameters

Parameter	Type	Default	Description
temperature	float	0.7	Controls randomness (0.0 - 2.0)
max_tokens	integer	2048	Maximum tokens in the response
top_p	float	1.0	Nucleus sampling parameter
frequency_penalty	float	0.0	Reduces repetition (-2.0 to 2.0)
presence_penalty	float	0.0	Encourages new topics (-2.0 to 2.0)
stop	string or array	null	Sequences where generation stops

Example Requests

Simple Conversation

import requests

# シンプルな会話リクエスト
response = requests.post(
    "https://api.rustellar.com/v1/chat/completions",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json"
    },
    json={
        "model": "helix-v1",
        "messages": [
            {"role": "user", "content": "Tell me a joke"}
        ]
    }
)

print(response.json())

Multi-turn Conversation

# 複数ターンの会話履歴を含むリクエスト
response = requests.post(
    "https://api.rustellar.com/v1/chat/completions",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json"
    },
    json={
        "model": "helix-v1",
        "messages": [
            {"role": "system", "content": "You are a helpful assistant."},
            {"role": "user", "content": "What is 2+2?"},
            {"role": "assistant", "content": "2+2 equals 4."},
            {"role": "user", "content": "What about 3+3?"}
        ],
        "temperature": 0.7,
        "max_tokens": 100
    }
)

print(response.json())

Story Generation with Iroha

# Irohaモデルでストーリー生成
response = requests.post(
    "https://api.rustellar.com/v1/chat/completions",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json"
    },
    json={
        "model": "iroha",
        "messages": [
            {
                "role": "system",
                "content": "You are a creative storyteller."
            },
            {
                "role": "user",
                "content": "Write a short story about a space explorer."
            }
        ],
        "temperature": 0.9,  # より創造的な出力のため高めに設定
        "max_tokens": 2000
    }
)

print(response.json())

Best Practices

1. Use system messages to set consistent behavior
2. Keep conversation history for context in multi-turn dialogues
3. Adjust temperature based on use case:
   • Lower (0.1-0.5) for factual, consistent responses
   • Higher (0.7-1.2) for creative content
4. Set appropriate max_tokens to control response length and costs
5. Use stop sequences to prevent unwanted continuation

Next Steps

• Learn about Output Formats
• Check Rate Limits
• Explore Model Options