Supported AI Providers
ReplyWolf works with a wide range of AI providers. Pick the one that fits your needs and budget.
Provider Overview
| Provider | Models | Get Your API Key |
|---|---|---|
| OpenAI | gpt-4o, gpt-4o-mini, gpt-4-turbo, gpt-3.5-turbo | platform.openai.com |
| Anthropic | Claude Sonnet 4.5, Claude Haiku 4.5, Claude Opus 4 | console.anthropic.com |
| Google Gemini | gemini-2.0-flash, gemini-2.0-flash-lite, gemini-1.5-flash | aistudio.google.com |
| Groq | llama-3.1-8b-instant, llama-3.3-70b-versatile, gemma2-9b-it | console.groq.com |
| DeepSeek | deepseek-chat, deepseek-reasoner | platform.deepseek.com |
| Kimi (Moonshot) | moonshot-v1-8k, moonshot-v1-32k, moonshot-v1-128k | platform.moonshot.cn |
| OpenRouter | 100+ models via one API | openrouter.ai |
| Custom | Any OpenAI-compatible API | Your provider's site |
Provider Details
OpenAI
The most popular choice. GPT-4o is their flagship model — fast, accurate, and great at following instructions. GPT-4o-mini is a cheaper alternative that still performs well for most tasks.
Anthropic
Claude models are known for being thoughtful and following nuanced instructions. Claude Sonnet 4.5 is a strong all-rounder. Claude Opus 4 is the most capable option.
Google Gemini
Gemini Flash models are fast and very affordable. A solid pick if you want to keep costs low without sacrificing too much quality.
Groq
Groq offers a free tier, making it a great option for testing or low-volume use. Their models (based on Llama and Gemma) run extremely fast thanks to Groq's custom hardware.
DeepSeek
DeepSeek offers competitive models at low prices. The deepseek-chat model handles general conversation well, while deepseek-reasoner is tuned for more complex reasoning.
Kimi (Moonshot)
Moonshot models come in different context window sizes (8k, 32k, 128k tokens). Choose a larger window if your flows need to process longer conversations.
OpenRouter
OpenRouter acts as a gateway to 100+ models from various providers through a single API key. Popular choices include openrouter/auto (automatically picks the best model), google/gemini, anthropic/claude, and openai/gpt-4o-mini.
Custom Provider
If you run your own AI server or use a provider not listed above, choose Custom and enter your base URL and API key. Any service that exposes an OpenAI-compatible API will work.
Which Provider Should I Use?
- Best quality: OpenAI (GPT-4o) or Anthropic (Claude Sonnet 4.5)
- Budget-friendly: Google Gemini Flash (very cheap) or Groq (free tier available)
- Maximum flexibility: OpenRouter (access many models with one key)