Moa

Availability

Moa is available now in paid Autohand Code plans and through the Autohand API. Use the exact model identifier moa.

Moa exposes medium, high, and xhigh reasoning settings in Autohand Code. Choose the smallest setting that gives the task enough time to reason before it acts.

Technical specifications

When to use Moa

Choose Moa when the answer depends on relationships across a repository, architecture decisions, or a plan that must account for more than a local change.

• Large refactors, migration plans, and multi-file changes
• Architecture work, repository analysis, and security review
• Tasks where the agent needs to compare patterns and constraints before editing

Choose Fantail for smaller follow-up edits and fast review cycles.

Use Moa

Autohand Code and the HTTP API both select Moa with the model ID. Keep that ID explicit in scripts and request logs so a run can be understood and reproduced later.

autohand --model moa --prompt "Map the authentication flow and propose the safest refactor plan." --patch

# In an active session
/model moa

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.AUTOHAND_API_KEY,
  baseURL: "https://api.autohand.ai/v1",
});

const response = await client.chat.completions.create({
  model: "moa",
  messages: [
    { role: "user", content: "Map the authentication flow and propose the safest refactor plan." },
  ],
  stream: true,
});

curl https://api.autohand.ai/v1/chat/completions \
  -H "Authorization: Bearer $AUTOHAND_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "moa",
    "messages": [{
      "role": "user",
      "content": "Map the authentication flow and propose the safest refactor plan."
    }]
  }'

Prompting for repository work

Start with the outcome and constraints. Tell the agent what it must inspect, which contracts must stay stable, and which checks define success. This gives the reasoning loop a concrete path through a large codebase.

Map the authentication flow across the service and web client. Propose the smallest safe migration, identify backwards-compatibility risks, then implement only after the plan names the tests that will verify it.

Plans and limits

Moa is available on Pro and above. The Free plan does not include Moa. Credits and request limits are set by the plan, not by the model, and are listed in full on the plans and pricing pages.

See plans and pricing for the full limits.

Capabilities and limits

The published contract covers chat completion requests, model selection, tool calling, and the available reasoning settings. This reference does not claim support for response formats, vision inputs, streaming controls, or sampling parameters. Treat those capabilities as unavailable until documented by the API contract.

Keep AUTOHAND_API_KEY on the server or in your local environment. Do not put API keys in browser code or source control.