vrr/eigent
2
0
Fork 0
mirror of https://github.com/eigent-ai/eigent.git synced 2026-05-17 04:01:18 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1
eigent/docs/core/models/byok.md
Douglas Lai 9a9d008bb6
Some checks failed
Pre-commit / pre-commit (push) Has been cancelled
Details
Test / Run Python Tests (push) Has been cancelled
Details
CodeQL Advanced / Analyze (actions) (push) Has been cancelled
Details
CodeQL Advanced / Analyze (javascript-typescript) (push) Has been cancelled
Details
CodeQL Advanced / Analyze (python) (push) Has been cancelled
Details
Codex/skill docs update - New PR with correct video formate (#1387) 
Co-authored-by: Tao Sun <168447269+fengju0213@users.noreply.github.com>
Co-authored-by: Sun Tao <2605127667@qq.com>
2026-02-27 21:24:58 +08:00

6.3 KiB
Raw Permalink Blame History

title description
Bring Your Own Key (BYOK) Configure your own API keys to use various LLM providers with Eigent.

What is BYOK?

Bring Your Own Key (BYOK) allows you to use your own API keys from various AI model providers with Eigent. Instead of relying on a shared service, you connect directly to providers like OpenAI, Anthropic, or Google using your personal API credentials. This gives you:

  • Full control over your API usage and billing
  • Direct access to the latest models from each provider
  • Privacy - your requests go directly to the provider

OpenAI Configuration (Example)

Step 1: Get Your API Key

  1. Visit the OpenAI API Keys page
  2. Click "Create new secret key"
  3. Copy the generated key (you won't be able to see it again)

Step 2: Configure in Eigent

  1. Launch Eigent and go to Agent > Models

  2. Find the OpenAI card in the Custom Model section

byok_1

  1. Fill in the following fields:
Field Value Example
API Key Your OpenAI secret key sk-proj-xxxx...
API Host OpenAI API endpoint https://api.openai.com/v1
Model Type The model you want to use gpt-4o, gpt-4o-mini
  1. Click Save to validate and store your configuration
  2. Click Set as Default to use this provider for your agents

Configuration Fields

Field Description Required
API Key Your authentication key from the provider Yes
API Host The API endpoint URL Yes (pre-filled for most providers)
Model Type The specific model variant to use Yes
External Config Provider-specific settings (e.g., Azure deployment name) Only for certain providers

Azure-Specific Fields

Field Description Example
API Version Azure OpenAI API version 2024-02-15-preview
Deployment Name Your Azure deployment name my-gpt4-deployment

Common Errors

When saving your configuration, Eigent validates your API key and model. Here are the errors you may encounter:

Error Cause Solution
Invalid key. Validation failed. API key is incorrect, expired, or malformed Double-check your API key. Regenerate a new key if needed.
Invalid model name. Validation failed. The specified model does not exist or is not available for your account Verify the model name is correct. Check if you have access to that model.
You exceeded your current quota API quota exhausted or billing issue Check your provider's billing dashboard. Add credits or upgrade your plan.

Supported Providers

Eigent supports the following BYOK providers:

Provider Default API Host Official Documentation
OpenAI https://api.openai.com/v1 OpenAI API Docs
Anthropic https://api.anthropic.com/ Anthropic API Docs
Google Gemini https://generativelanguage.googleapis.com/v1beta/openai/ Gemini API Docs
OpenRouter https://openrouter.ai/api/v1 OpenRouter Docs
Qwen (Alibaba) https://dashscope.aliyuncs.com/compatible-mode/v1 Qwen API Docs
DeepSeek https://api.deepseek.com DeepSeek API Docs
Minimax https://api.minimax.io/v1 Minimax API Docs
Z.ai https://api.z.ai/api/coding/paas/v4/ Z.ai Platform
Azure OpenAI (user-provided) Azure OpenAI Docs
AWS Bedrock (user-provided) AWS Bedrock Docs
OpenAI Compatible (user-provided) For custom endpoints (e.g., xAI, local servers)

Tips

  • Keep your API key secure - Never share or expose your API key publicly
  • Monitor usage - Check your provider's dashboard regularly to track costs
  • Use appropriate models - Different models have different capabilities and pricing