Back to docs

API Key Configuration Guide

2026-05-02

API Key Configuration Guide

This guide teaches you how to configure AI models in Doforu, just like filling out a simple form.

1. Getting to Know the Settings Page

Click the Settings button in the bottom-left corner of the Doforu interface. The settings window will pop up, and by default, you're on the API Key page.

Click the Settings button in the bottom-left corner of the Doforu interface. The settings window will pop up, and by default, you're on the API Key page.

API Key configuration page showing provider options

API Key configuration with image3

2. Supported Model Providers

Doforu supports OpenAI Compatible and Anthropic Native protocols. Below are all supported AI providers.

OpenAI Compatible Protocol

Works with most AI providers. Select Provider: OpenAI Compatible.

Provider Base URL Models Cost Region Best For Get Key
OpenAI • GPT-5.5 api.openai.com/v1 gpt-5.5 $5 / $30 US Flagship, strongest model Get Key
OpenAI • GPT-5.4 api.openai.com/v1 gpt-5.4 $2.50 / $15 US Production workhorse, 1M context Get Key
OpenAI • GPT-5.4-Mini api.openai.com/v1 gpt-5.4-mini $0.75 / $4.50 US Budget-friendly general tasks Get Key
OpenAI • o4-mini api.openai.com/v1 o4-mini $0.55 / $2.20 US Budget reasoning, math, logic Get Key
OpenAI • GPT-4.1 api.openai.com/v1 gpt-4.1 $2 / $8 US Reliable long-context production model Get Key
Google Gemini • 2.5 Flash generativelanguage.googleapis.com/v1beta/openai/ gemini-2.5-flash $0.30 / $2.50 US Fast hybrid reasoning, 1M context Get Key
Google Gemini • 2.5 Flash-Lite generativelanguage.googleapis.com/v1beta/openai/ gemini-2.5-flash-lite $0.10 / $0.40 US Most cost-efficient, high volume Get Key
Google Gemini • 3 Flash Preview generativelanguage.googleapis.com/v1beta/openai/ gemini-3-flash-preview $0.50 / $3.00 US Frontier intelligence, fast Get Key
Google Gemini • 3.1 Pro Preview generativelanguage.googleapis.com/v1beta/openai/ gemini-3.1-pro-preview $2 / $12 US Best multimodal, agentic, vibe-coding Get Key
DeepSeek • V4 Flash api.deepseek.com deepseek-v4-flash $0.14 / $0.28 CN Agentic coding, low cost Get Key
DeepSeek • V4 Pro api.deepseek.com deepseek-v4-pro $1.74 / $3.48 CN Maximum capability Get Key
DeepSeek • V3 api.deepseek.com deepseek-v3.2 / deepseek-v3 / deepseek-chat $0.14 / $0.28 CN Stable legacy models, great value Get Key
Anthropic • Claude Sonnet 4.6 api.anthropic.com/v1 claude-sonnet-4.6 / claude-3-5-sonnet-20241022 $3 / $15 US Best balance for production Get Key
Anthropic • Claude Opus 4.7/4.6 api.anthropic.com/v1 claude-opus-4.7 / claude-opus-4.6 $5 / $25 US Best conversation, long context Get Key
Anthropic • Claude Haiku 4.5 api.anthropic.com/v1 claude-haiku-4.5 $1 / $5 US Fast, lightweight tasks Get Key
OpenRouter openrouter.ai/api/v1 owl-alpha / openai/gpt-5.5 / openai/gpt-5.4 / anthropic/claude-opus-4.7 / anthropic/claude-sonnet-4.6 $2.50 / $10 US Aggregates multiple models Get Key
xAI (Grok) api.x.ai/v1 grok-4.3 / grok-4 / grok-4.1-fast $1.25 / $2.50 US Fast reasoning, competitive pricing Get Key
TogetherAI api.together.xyz/v1 meta-llama/Llama-3.3-70B-Instruct / mistralai/Mixtral-8x22B $0.6 / $1.2 US Open source models cloud hosting Get Key
Groq api.groq.com/openai/v1 llama-3.1-70b-versatile / mixtral-8x7b-32768 / nemotron-3-super $0.2 / $0.4 US Ultra-low latency inference (LPU chip) Get Key
Tencent Hunyuan api.hunyuan.cloud.tencent.com/v1 hunyuan-turbos-latest / hunyuan-pro / hunyuan-lite $0.4 / $1.2 CN Tencent ecosystem, strong Chinese language Get Key
StepFun api.stepfun.com/v1 step-3.5-flash $0.3 / $0.6 CN Chinese ecosystem, stable Get Key
MiniMax api.minimax.chat/v1 minimax-m2.7 $0.2 / $0.4 CN Affordable, good performance Get Key
SiliconFlow api.siliconflow.cn/v1 Qwen/Qwen3.5-35B-A3B / DeepSeek-V3.2 / DeepSeek-V4 $0.99 / $1.99 CN Works in China, free credits Get Key
Alibaba Bailian dashscope.aliyuncs.com/compatible-mode/v1 qwen3.7-max / qwen-max / qwen-plus $0.5 / $2 CN Alibaba ecosystem, stable Get Key
Zhipu AI open.bigmodel.cn/api/paas/v4 glm-5.1 / glm-4 / glm-4-flash $0.1 / $0.3 CN Works in China, optimized Chinese Get Key
Moonshot Kimi api.moonshot.cn/v1 moonshot-k2.6 / moonshot-v1-128k $0.6 / $1.2 CN Long context support Get Key
ByteDance Doubao ark.cn-beijing.volces.com/api/v3 doubao-seed-2.0-pro / doubao-pro / doubao-lite $0.2 / $1 CN Affordable and stable Get Key

Anthropic Native Protocol

For Claude models. Select Provider: Anthropic Native.

Provider Base URL Models Cost Region Best For Get Key
Anthropic • Claude Sonnet 4.6 api.anthropic.com/v1 claude-sonnet-4.6 / claude-3-5-sonnet-20241022 $3 / $15 US Best balance for production Get Key
Anthropic • Claude Opus 4.7 api.anthropic.com/v1 claude-opus-4.7 $5 / $25 US Best conversation, long context Get Key
Anthropic • Claude Haiku 4.5 api.anthropic.com/v1 claude-haiku-4.5 $1 / $5 US Fast, lightweight tasks Get Key

Note: Anthropic Native protocol does NOT require "OpenAI Compatible" checkbox to be enabled.

OpenRouter (Unified Multi-Model Gateway)

OpenRouter acts as a single API proxy that gives you access to 200+ models from dozens of providers—OpenAI, Anthropic, Google, Meta, DeepSeek, and more—all through one API key. Instead of juggling multiple provider accounts, you configure once and switch models by changing the model ID.

How model IDs work: OpenRouter uses provider/model-name format. For example:

  • openai/gpt-5.5 — OpenAI's GPT-5.5
  • anthropic/claude-sonnet-4.6 — Anthropic's Claude Sonnet 4.6
  • google/gemini-2.5-flash — Google's Gemini 2.5 Flash
  • openai/gpt-5.4 — OpenAI's GPT-5.4
  • anthropic/claude-opus-4.7 — Anthropic's Claude Opus 4.7
  • meta-llama/llama-3.3-70b-instruct — Meta's Llama 3.3 (via OpenRouter)

Browse all available models at openrouter.ai/models.

Quick setup (4 steps):

  1. Sign up at openrouter.ai and get your API key
  2. In Doforu, add a new config with Provider: OpenAI Compatible
  3. Set Base URL to https://openrouter.ai/api/v1
  4. Enter your desired Model ID (e.g., openai/gpt-5.5) and your API key

Why choose OpenRouter? One key, one base URL, access to 200+ models. Great for exploring different models without creating multiple provider accounts.

3. Configuration Steps

Step 1: Open Settings

Click the Settings button in the bottom-left corner. Settings modal opens (default: API Key page).

Step 2: Add New Configuration

Click + Add New Config in the top-right corner.

Step 3: Select Provider Protocol

  • OpenAI-based models → Select OpenAI Compatible
  • Claude models → Select Anthropic Native

Step 4: Fill in Configuration

Copy the corresponding Base URL and Model from the tables above.

Step 5: Paste API Key

Copy your API Key from the provider's website and paste it into the input box.

Step 6: Save Configuration

  1. Click Test Connection → Wait 2-3 seconds
  2. Shows green success → Click Save
  3. Your model will appear in the model selector at the top-right

4. Configuration Templates

Template A: OpenAI Compatible

Field Value
Provider OpenAI Compatible
Base URL Provider's Base URL (e.g., https://api.openai.com/v1)
Model Provider's example model (e.g., gpt-5.5)
API Key Your API Key from the provider
Config Name Any name (e.g., "My GPT-5.5")

Template B: Anthropic Native (Claude only)

Field Value
Provider Anthropic Native
OpenAI Compatible Do NOT enable
Base URL https://api.anthropic.com/v1
Model claude-sonnet-4.6
API Key Your API Key from Anthropic
Config Name Any name (e.g., "My Claude")

Template C: OpenRouter (Multi-Provider Gateway)

Field Value
Provider OpenAI Compatible
Base URL https://openrouter.ai/api/v1
Model Provider's model ID (e.g., openai/gpt-5.5)
API Key Your API Key from openrouter.ai/keys
Config Name Any name (e.g., "OpenRouter GPT-5.5")

5. Connection Test and Troubleshooting


Connection Test dialog showing test results


Test Results

Result Meaning Solution
Green "Connection Success" Config is correct Save and use
Red "Connection Failed" Config has errors Check the following

Troubleshooting

1. Incorrect Base URL

  • Make sure it includes the https:// prefix
  • Check for extra/missing slashes /

2. Incorrect API Key

  • Make sure you copied it completely (no extra spaces)
  • Check if the key is expired or disabled

3. Network Issues

  • Some providers require VPN to access
  • Check if your proxy is blocking requests

4. Incorrect Model Name

  • Confirm the model name matches exactly
  • Some providers include version numbers (e.g., gpt-5.5-2026-04-24)

6. FAQ

Q: Can I configure multiple providers at the same time?

A: Yes. Click "Add New Config" to add multiple providers, then switch between them in the model selector.

Q: Which provider is best for beginners?

A: Users in China: SiliconFlow (free credits, works immediately). International users: OpenAI (most popular, strongest models).

Q: Is my API Key secure?

A: Your API Key is stored locally or on your configured server. Doforu does not collect or upload your Key.

Q: How do I choose a model?

A: Recommended starting models:

  • gpt-5.5 / gpt-5.4-mini (OpenAI)
  • deepseek-chat (DeepSeek - lowest cost)
  • glm-4-flash (Zhipu AI - lots of free credits in China)

Q: Should I use OpenRouter or configure each provider separately?

A: If you want to try many different models quickly, start with OpenRouter—one key, one setup, access to 200+ models. If you have heavy production usage and want the absolute lowest cost for a specific provider, configuring that provider directly may be cheaper.