Magnum v4 72B

Description

Magnum v4 72B is a fine-tuned version of Qwen2.5 72B that aims to replicate the prose quality of Claude 3 models, specifically Sonnet and Opus. This model is designed for creative writing and roleplay scenarios, offering high-quality text generation with a focus on narrative and conversational abilities.

The model uses the ChatML instruction format and is optimized for generating coherent, engaging prose that matches the stylistic qualities of Anthropic's Claude 3 family.

Technical Specifications

Pricing

• Input: $0.000003 per token
• Output: $0.000005 per token

Supported Parameters

Use Cases

• Creative Writing: Novel writing, short stories, poetry
• Roleplay: Character-driven conversations and scenarios
• Dialogue Generation: Natural, engaging conversations
• Narrative Content: Blog posts, articles with storytelling elements
• Interactive Fiction: Text-based games and adventures

Limitations

• No vision/image input support
• No tool/function calling support
• No reasoning/chain-of-thought capabilities
• Context limited to 16,384 tokens (vs 32,768 in base model)
• Maximum completion limited to 2,048 tokens

Providers

Model Information

Feature Support

Default Stop Tokens

<|im_start|>
<|im_end|>
<|endoftext|>

Usage Examples

LangMart API (Python)

import requests

response = requests.post(
    "https://api.langmart.ai/v1/chat/completions",
    headers={
        "Authorization": "Bearer YOUR_LANGMART_API_KEY",
        "Content-Type": "application/json"
    },
    json={
        "model": "anthracite-org/magnum-v4-72b",
        "messages": [
            {
                "role": "user",
                "content": "Write a short story about a mysterious lighthouse."
            }
        ],
        "max_tokens": 1024,
        "temperature": 0.8
    }
)

print(response.json()["choices"][0]["message"]["content"])

LangMart API (cURL)

curl https://api.langmart.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_LANGMART_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthracite-org/magnum-v4-72b",
    "messages": [
      {
        "role": "user",
        "content": "Write a short story about a mysterious lighthouse."
      }
    ],
    "max_tokens": 1024,
    "temperature": 0.8
  }'

LangMart API (JavaScript/Node.js)

const response = await fetch("https://api.langmart.ai/v1/chat/completions", {
  method: "POST",
  headers: {
    "Authorization": "Bearer YOUR_LANGMART_API_KEY",
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    model: "anthracite-org/magnum-v4-72b",
    messages: [
      {
        role: "user",
        content: "Write a short story about a mysterious lighthouse."
      }
    ],
    max_tokens: 1024,
    temperature: 0.8
  })
});

const data = await response.json();
console.log(data.choices[0].message.content);

ChatML Format Example

<|im_start|>system
You are a creative writing assistant specialized in crafting engaging narratives.
<|im_end|>
<|im_start|>user
Write a short story about a mysterious lighthouse.
<|im_end|>
<|im_start|>assistant

Recommended Settings

For creative writing and roleplay:

For more deterministic outputs:

Source

• Documentation: https://langmart.ai/model-docs
• Base Model: Qwen2.5 72B by Alibaba Cloud