API Key Management

API Keys

TalkCody supports multiple AI providers, each requiring their own API key:

• OpenAI - GPT models
• Anthropic - Claude models
• Google - Gemini models
• OpenRouter - Gateway supporting 500+ models
• Vercel AI Gateway - Gateway supporting 100+ models
• Alibaba Qwen - Qwen models
• Zhipu AI - GLM models
• Z.AI - GLM models (International)
• MiniMax - MiniMax models
• DeepSeek - DeepSeek models
• Moonshot - Kimi models
• Ollama - Local models (no key needed)
• LM Studio - Local models (no key needed)

Adding API Keys

Getting API Keys

Vercel AI Gateway

Multi-model AI Gateway that provides access to multiple AI provider models through a single interface.

Steps to obtain:

1. Visit Vercel AI Gateway
2. Sign in or register with your Vercel account
3. Create a new project in the Dashboard
4. Go to project settings and find AI Gateway configuration
5. Generate a Gateway API Key
6. Copy the key and paste it into TalkCody's Vercel AI Gateway field

OpenRouter

OpenRouter API Gateway that aggregates multiple AI providers, providing access to 100+ models through a single API.

Steps to obtain:

1. Visit openrouter.ai
2. Click "Sign In" in the top right corner to login or register
3. After logging in, click your avatar to enter account settings
4. Select "Keys" from the left menu
5. Click "Create Key" to create a new key
6. Name your key (e.g., "TalkCody") and click confirm
7. Copy the generated key (format: sk-or-v1-...)
8. Paste it into TalkCody's OpenRouter field

OpenAI

Official OpenAI API, providing advanced language models like GPT-4 and GPT-4 Turbo.

Steps to obtain:

1. Visit platform.openai.com
2. Click "Sign up" to register or "Log in" to sign in
3. Complete account verification (may require phone number)
4. After logging in, click your avatar in the top right → "API keys"
5. Click "Create new secret key"
6. Name your key and click "Create secret key"
7. Copy the key immediately (format: sk-..., cannot be viewed again after closing)
8. Paste it into TalkCody's OpenAI field

Custom Base URL: If using a proxy server or OpenAI-compatible third-party service (like MiniMax), enter the corresponding endpoint address in the "Custom Base URL" field.

Anthropic

Official Anthropic Claude API, providing Claude 3.5 Sonnet, Opus, Haiku and other models.

Steps to obtain:

1. Visit console.anthropic.com
2. Click "Sign up" to register a new account
3. Complete email verification and account setup
4. After logging in, enter the Console Dashboard
5. Click "API Keys" in the left menu
6. Click the "Create Key" button
7. Enter a key name and confirm creation
8. Copy the generated key (format: sk-ant-...)
9. Paste it into TalkCody's Anthropic field

Custom Base URL: If you need to use a proxy server, you can configure a custom endpoint in the "Custom Base URL" field.

Google AI

Google AI Studio's Gemini series models, supporting text and multimodal capabilities.

Steps to obtain:

1. Visit aistudio.google.com
2. Sign in with your Google account
3. Agree to the terms of service
4. Click "Get API key" in the left menu
5. Click "Create API key in new project" or select an existing project
6. Copy the generated key (format: AIza...)
7. Paste it into TalkCody's Google AI field

MiniMax

MiniMax's large language model service, supporting text generation and conversation capabilities.

Steps to obtain:

1. Visit platform.minimaxi.com
2. Click register or login in the top right corner
3. Complete account registration and verification
4. After logging in, enter the user center
5. In the left menu, select "Basic Information" → "Interface Key"
6. Or directly visit the Interface Key page
7. Click "Create Key" or view existing keys
8. Copy the API Key
9. Paste it into TalkCody's MiniMax field

Zhipu AI

Zhipu AI's GLM series models, including GLM-4 and other Chinese-optimized models.

Steps to obtain:

1. Visit open.bigmodel.cn
2. Click "Register" to create an account (supports phone number registration)
3. Complete identity verification (may be required)
4. After logging in, enter the console
5. Go to the "API Key Management" page
6. Click "Create API Key"
7. Copy the generated key
8. Paste it into TalkCody's Zhipu AI field

Ollama

Locally running open-source models, completely offline, privacy-focused.

Configuration steps:

1. Visit ollama.com to download and install Ollama
2. Open terminal and download a model:

   ollama pull llama2
   # or other models like codellama, mistral, etc.

3. Ensure Ollama service is running
4. TalkCody will automatically detect local Ollama
5. No API key needed, directly select Ollama models in model settings

DeepSeek

DeepSeek's high-performance large language models, offering powerful code generation and reasoning capabilities at highly competitive prices.

Steps to obtain:

1. Visit platform.deepseek.com
2. Click "Sign Up" or "Login" in the top right corner
3. Complete email verification and account setup
4. After logging in, enter the console
5. Select "API Keys" from the left menu
6. Click "Create new API key"
7. Name your key and confirm
8. Copy the generated key (format: sk-...)
9. Paste it into TalkCody's DeepSeek field

Moonshot

Moonshot's Kimi large language model, featuring ultra-long context windows, supporting long document processing and complex conversations.

Steps to obtain:

1. Visit platform.moonshot.cn
2. Click "Login" or "Register" in the top right corner
3. Complete registration and verification with phone number
4. After logging in, enter the console
5. Select "API Key Management" from the left menu
6. Click "New" to create a new key
7. Name your key and confirm
8. Copy the generated key (format: sk-...)
9. Paste it into TalkCody's Moonshot field

LM Studio

LM Studio is a desktop application that lets you run open-source large language models locally, providing a graphical interface to manage and run models.

Configuration steps:

1. Visit lmstudio.ai to download and install LM Studio
2. Open the LM Studio application
3. Search and download your desired model in the "Discover" tab
4. After download completes, go to the "Developer" tab
5. Click "Start Server" to start the local server
6. Server runs by default at http://localhost:1234
7. No API key needed in TalkCody, simply select LM Studio models

Tavily

Search API designed specifically for AI agents and large language models, providing accurate, real-time web search results.

Steps to obtain:

1. Visit tavily.com
2. Click "Get API Key" or "Sign Up"
3. Register with your email
4. Complete email verification
5. After logging in, enter the Dashboard
6. View your key in the "API Keys" section
7. Copy the API key
8. Paste it into TalkCody's Tavily field

Eleven Labs

Leading AI voice synthesis platform, providing high-quality text-to-speech and voice cloning features.

Steps to obtain:

1. Visit elevenlabs.io
2. Click "Sign Up" in the top right corner
3. Register with Google account or email
4. Complete account creation and verification
5. After logging in, click your avatar in the top right
6. Select "Profile + API key"
7. Click the eye icon in the API Key section to view the key
8. Copy the API key (format: long string of numbers and letters)
9. Paste it into TalkCody's Eleven Labs field

Serper

Google Search API service that provides fast and accurate web search results, designed for AI applications and automation.

Steps to obtain:

1. Visit serper.dev
2. Click "Get API Key" or "Sign Up"
3. Register with your email or Google account
4. Complete email verification
5. After logging in, enter the Dashboard
6. View your API key in the dashboard
7. Copy the API key
8. Paste it into TalkCody's Serper field

Z.AI

Z.AI provides advanced language models from the GLM family, including high-performance models like GLM-4.7.

Steps to obtain:

1. Visit z.ai/model-api
2. Click "Sign Up" to register an account or "Login" to sign in
3. Complete email verification and account setup
4. After logging in, enter the console
5. Select "API Keys" from the left menu
6. Click "Create API Key"
7. Copy the generated key
8. Paste it into TalkCody's Z.AI field