Skip to content

Latest commit

 

History

History
241 lines (209 loc) · 6.37 KB

axon-code.mdx

File metadata and controls

241 lines (209 loc) · 6.37 KB
title description og:title og:description og:image og:url
Axon Code 1
Super Intelligent model for coding with deep reasoning
Axon Code Model By MatterAI | MatterAI Documentation
Axon Code Model By MatterAI is a super intelligent model for coding with deep reasoning

Axon Code

Specification Value
ModelID axon-code
Description Code-gen LLM Model with deep-reasoning
Region US
Context Window Size 256K tokens
Max Output Tokens 32,768
Input Price (<256K) $1.0/1M tokens
Output Price (<256K) $4.0/1M tokens
Input Modalities Text, Image
Output Modalities Text
Capabilities Function Calling, Tool Calling, Reasoning
Parameters 480B
Floating Point Precision FP16
Axon Code is based on Qwen 3 480B model, fine tuned on our proprietary dataset and upgraded with deep reasoning, code generation and state machine capabilities.

Key Features

  • Production Code: Enterprise-grade, deployable code
  • Security: OWASP compliance, vulnerability detection
  • Organizational Learning: Adapts to codebase patterns
  • Data Privacy: Client data isolation
  • Platform Integration: Jira, GitHub, GitLab connectivity
  • Deep Reasoner: Deep Reasoner Engine with search and web fetch tool calling
  • State Machine: Manages complex workflows and transitions

Benchmarks

API & SDK Integration

curl --request POST \
  --url https://api.matterai.so/v1/chat/completions \
  --header 'Content-Type: application/json' \
  --header 'Authorization: Bearer MATTER_API_KEY' \
  --data '{
  "model": "axon-code",
  "messages": [
    {
      "role": "system",
      "content": "You are a helpful assistant."
    },
    {
      "role": "user",
      "content": "Explain Haskell paradigms"
    }
  ],
  "stream": false,
  "max_tokens": 1000,
  "reasoning": {
    "effort": "high",
    "summary": "none"
  },
  "response_format": {
    "type": "text"
  },
  "temperature": 0,
  "top_p": 1
}'
import OpenAI from "openai";

const openai = new OpenAI({
  apiKey: "MATTER_API_KEY",
  baseURL: "https://api.matterai.so/v1",
});

async function main() {
  const response = await openai.chat.completions.create({
    model: "axon-code",
    messages: [
      {
        role: "system",
        content: "You are a helpful assistant.",
      },
      {
        role: "user",
        content: "What is Rust?",
      },
    ],
    stream: false,
    max_tokens: 1000,
    reasoning: {
      effort: "high",
      summary: "none",
    },
    response_format: {
      type: "text",
    },
    temperature: 0,
    top_p: 1,
  });

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

main();
from openai import OpenAI

client = OpenAI(
  api_key='MATTER_API_KEY',
  base_url='https://api.matterai.so/v1'
)

response = client.chat.completions.create(
  model='axon-code',
  messages=[
    {
      'role': 'system',
      'content': 'You are a helpful assistant.'
    },
    {
      'role': 'user',
      'content': 'What is Rust?'
    }
  ],
  stream=False,
  max_tokens=1000,
  reasoning={
    'effort': 'high',
    'summary': 'none'
  },
  response_format={
    'type': 'text'
  },
  temperature=0,
  top_p=1
)

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

IDE Integrations

Integrate Axon Code with Cursor Integrate Axon Code with Cline

Axon Intelligence

Code Quality

Axon Code recognizes organizational coding patterns across the codebase to generate code that is consistent with the organization's standards. Axon Code understands and removes as much as possible anti-patterns in it's generated code, powered by out deep reasoner, it can pre-detect where anti-patterns can occur.

Security & Performance

Axon Code is well versed with Static Application Security Testing (SAST) to pre-determine potential security vulnerabilities, there by generating code that is secure. Axon Code generates sleek, performant and optimized code that is always fast to run for your CPU cycles and RAM usage.

Code Maintainability

Axon Code only writes code that matters, there by reducing the amount of code you need to generate to get the job done. Axon Code generates detailed documentation for your code, ensuring that your code is easy to understand and maintain.

Model Family

Learn about the core technologies and general integration of Axon models.