Custom Models & Providers

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/21st-dev/1code/llms.txt

Use this file to discover all available pages before exploring further.

​

Overview

• Claude Subscription - OAuth with your Anthropic account (Claude Pro/Max)
• Custom Models - Use custom base URLs and API keys for proxies or alternative providers
• Ollama - Local models for offline development

​

Using Custom Model Providers

• Use API proxies or alternative Anthropic-compatible endpoints
• Connect to custom deployments with your own API keys
• Configure different base URLs per project or globally

​

Configuration Options

​

Environment Variables

export ANTHROPIC_API_KEY="your-api-key"
export ANTHROPIC_BASE_URL="https://your-proxy.com"

​

How It Works

1. 1Code validates the configuration by checking model availability
2. The custom base URL is used for all API requests
3. Your API key is encrypted using Electron’s safeStorage and stored locally
4. Environment variables are passed to the Claude CLI with custom settings

ANTHROPIC_AUTH_TOKEN: finalCustomConfig.token,
ANTHROPIC_BASE_URL: finalCustomConfig.baseUrl,

​

Multi-Account Management

​

Adding Accounts

1. Complete OAuth flow for a new account
2. Accounts are stored with encrypted tokens
3. Switch between accounts from Settings

​

Account Storage

• Email and display name
• Encrypted OAuth token (using Electron safeStorage)
• Connection and last-used timestamps
• Desktop user ID for sync

function encryptToken(token: string): string {
  if (!safeStorage.isEncryptionAvailable()) {
    console.warn("[AnthropicAccounts] Encryption not available, storing as base64")
    return Buffer.from(token).toString("base64")
  }
  return safeStorage.encryptString(token).toString("base64")
}

​

Ollama for Offline Development

​

Setup

1. Install Ollama from ollama.ai
2. Pull a model: ollama pull qwen2.5-coder
3. Configure Ollama in Settings

​

Offline Features

• Chat title generation - Uses local models for naming conversations
• Commit message generation - Generates commit messages without internet
• Model selection - Choose from installed Ollama models

async function generateWithOllama(
  prompt: string,
  model?: string | null
): Promise<string | null> {
  const ollamaStatus = await checkOllamaStatus()
  if (!ollamaStatus.available) {
    return null
  }

  const modelToUse = model || ollamaStatus.recommendedModel || ollamaStatus.models[0]
  
  const response = await fetch("http://localhost:11434/api/generate", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({
      model: modelToUse,
      prompt,
      stream: false,
      options: {
        temperature: 0.3,
        num_predict: 50,
      },
    }),
  })
}

​

Default Configuration

• Base URL: http://localhost:11434
• Recommended models: qwen2.5-coder:32b, qwen2.5-coder:14b, qwen2.5-coder:7b

​

Security Best Practices

​

Secure Token Storage

• Encryption at rest - Tokens are encrypted before storage
• OS-level security - Uses system keychain on macOS, Windows Credential Store, or libsecret on Linux
• Fallback mode - Base64 encoding if hardware encryption unavailable

​

Environment Variable Security

1. Add them to your shell profile (~/.zshrc, ~/.bashrc)
2. Never use .env files tracked by git
3. Use .env.local for local development (add to .gitignore)

​

API Key Rotation

1. Generate new API keys regularly
2. Revoke old keys from your provider dashboard
3. Update keys in Settings or environment variables

​

Troubleshooting

​

Custom Provider Not Working

curl https://your-proxy.com/v1/messages \
  -H "x-api-key: YOUR_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{"model": "claude-3-7-sonnet-20250219", "max_tokens": 10, "messages": [{"role": "user", "content": "Hello"}]}'

​

Ollama Not Detected

1. Verify Ollama is running: ollama list
2. Check the default port: curl http://localhost:11434/api/tags
3. Restart Ollama service if needed

​

Environment Variables Not Loading

1. User-configured settings (highest priority)
2. Vite env vars (.env.local files with MAIN_VITE_ prefix)
3. process.env
4. Shell environment (loaded via shell profile)

# Voice Input (optional - uses OpenAI Whisper API)
# Set this to enable voice-to-text for users without a paid subscription
# Get your API key at https://platform.openai.com/api-keys
# MAIN_VITE_OPENAI_API_KEY=sk-...

​

Advanced: Custom Provider Examples

​

AWS Bedrock Proxy

export ANTHROPIC_BASE_URL="https://bedrock-runtime.us-east-1.amazonaws.com"
export ANTHROPIC_API_KEY="your-aws-key"

​

Azure OpenAI Proxy

export ANTHROPIC_BASE_URL="https://your-resource.openai.azure.com"
export ANTHROPIC_API_KEY="your-azure-key"

​

Self-hosted Proxy

export ANTHROPIC_BASE_URL="https://api.yourcompany.com/anthropic"
export ANTHROPIC_API_KEY="your-internal-key"

​

Related Resources