Skip to main content

​
Connecting to External LLM Providers

ByteRover CLI supports connecting to external LLM providers, giving you access to hundreds of models beyond the built-in ByteRover LLM. Currently supports OpenRouter with 200+ models, with more providers coming soon.
This feature is in BETA. API keys are stored securely in your system keychain.

​
Supported Providers

ProviderDescriptionAPI KeyStatus
ByteRoverInternal ByteRover LLM (default)NoAvailable
OpenRouterAccess 200+ models from multiple providersYesAvailable
AnthropicDirect Claude API accessYesComing Soon
OpenAIDirect GPT API accessYesComing Soon
GoogleGemini modelsYesComing Soon
GroqFast inference modelsYesComing Soon

​
Quick Start

Connect to OpenRouter and select a model in under a minute:
1

Open provider selection

In the Console tab, run:
/provider
2

Select OpenRouter

Choose OpenRouter from the provider list.Provider selection
3

Enter your API key

Get your key at openrouter.ai/keys and paste it when prompted.API key prompt
4

Select a model

Browse and select from 200+ available models:
/model
Model browser

​
The /provider Command

Connect to or switch between LLM providers. Aliases: /providers, /connect

​
Behavior

  1. Displays a selection prompt with all available providers
  2. Shows connection status indicators:
    • (Current) - The active provider
    • [Connected] - Previously connected, API key stored
  3. If selecting an unconnected provider that requires an API key:
    • Prompts for the API key
    • Validates the key against the provider’s API
    • Stores the key securely in your system keychain
  4. Sets the selected provider as active

​
Examples

# Open provider selection dialog
/provider

# Also accessible via aliases
/providers
/connect

​
The /model Command

Browse and select models from your active provider. Aliases: /models
Requires an external provider to be connected. If using ByteRover (default), this command will prompt you to switch to an external provider.

​
Model Information Displayed

  • Model name and provider
  • Pricing (input/output cost per million tokens)
  • Context window size (e.g., “128K ctx”, “1.0M ctx”)
  • Status indicators:
    • (Current) - Currently selected model
    • ★ - Favorited model
    • [Free] - Free to use

​
Sorting Order

  1. Favorite models (starred)
  2. Recently used models
  3. Alphabetically by provider, then model name
Models are cached for 1 hour to improve performance. The cache refreshes automatically when needed.

​
Favorites and Recent Models

ByteRover tracks your model preferences to make selection faster.

​
Favorite Models

Mark models as favorites to pin them at the top of the selection list. Favorites are indicated with a ★ symbol.

​
Recent Models

Your 10 most recently used models appear after favorites in the selection list, making it easy to switch between frequently used models.

​
Configuration

​
Storage Locations

DataLocationPurpose
Provider config~/.config/brv/providers.jsonActive provider, favorites, recent models
API keysSystem keychainSecure credential storage

​
API Key Security

API keys are stored in your operating system’s secure keychain:

macOS

Keychain Access

Windows

Credential Manager

Linux

Secret Service (or encrypted file)

​
Configuration Structure

The provider configuration file stores non-sensitive preferences: 
{
  "activeProvider": "openrouter",
  "providers": {
    "openrouter": {
      "activeModel": "anthropic/claude-3.5-sonnet",
      "connectedAt": "2025-01-31T00:00:00Z",
      "favoriteModels": ["anthropic/claude-3.5-sonnet"],
      "recentModels": ["anthropic/claude-3.5-sonnet", "openai/gpt-4o"]
    }
  }
}

​
OpenRouter Setup

OpenRouter provides unified access to 200+ models from multiple providers through a single API key.

​
Getting an API Key

  1. Visit openrouter.ai/keys
  2. Sign up or log in
  3. Create a new API key
  4. Copy the key (starts with sk-or-...)
ModelProviderBest ForPricing (Input/Output)
anthropic/claude-3.5-sonnetAnthropicCoding, reasoning3.00/3.00 / 15.00 per M tokens
anthropic/claude-3.5-haikuAnthropicFast responses1.00/1.00 / 5.00 per M tokens
openai/gpt-4oOpenAIMultimodal tasks2.50/2.50 / 10.00 per M tokens
google/gemini-2.0-flashGoogleSpeed + quality balance0.10/0.10 / 0.40 per M tokens
meta-llama/llama-3.1-70bMetaOpen-source alternative0.35/0.35 / 0.40 per M tokens
Free models are marked with [Free] in the model selector. Great for testing and development.

​
Troubleshooting

Common causes:
  • API key is expired or revoked
  • Insufficient credits in your provider account
  • Network connectivity issues
Solution: Verify your API key at openrouter.ai/keys and ensure you have available credits.
Cause: The active provider requires an API key that hasn’t been configured.Solution: Run /provider to connect the provider with your API key first.
Run /provider and select ByteRover from the list. No API key is required for the built-in provider.
API keys are stored in your system keychain for security:
  • macOS: Keychain Access app (search for “byterover-cli-providers”)
  • Windows: Credential Manager
  • Linux: Your secret service manager

​
Next Steps

CLI Reference

Complete command reference including /provider and /model

Quickstart

Get started with ByteRover CLI