Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.openori.dev/llms.txt

Use this file to discover all available pages before exploring further.

OpenOri supports four model providers. You can configure multiple and switch between them anytime.

Supported providers

ProviderPopular modelsStrengths
AnthropicClaude 4, Claude SonnetBest reasoning, coding, tool use
OpenAIGPT-4.5, GPT-4oStrong general purpose, vision
GoogleGemini 2.5 Pro, FlashHuge context windows, multimodal
OllamaLlama 3, Mistral, PhiFull privacy, no API key, free

Adding a provider

Open Settings (⌘ ,) → scroll to Providers → enter your API key → press Enter or click Save.The provider is immediately enabled.
Use caseModelWhy
Daily driveranthropic:claude-sonnet-4-6Best balance of speed + capability
Complex reasoninganthropic:claude-opus-4-7Most capable Claude tier
Fast responsesopenai:gpt-4oQuick, solid for simple tasks
Large filesgoogle:gemini-2.5-pro1M token context window
Full privacyollama:llama3.1Everything runs locally

Switching models

Use the model selector at the bottom of the chat input. You can change models mid-conversation — no restart needed.

API key security

API keys are stored in your operating system’s native credential manager:
PlatformStorage
macOSKeychain Access
WindowsWindows Credential Manager
LinuxSecret Service (GNOME Keyring / KDE Wallet)
Keys never touch the filesystem in plaintext. They survive app reinstalls and ~/.ori/ deletion.

Getting API keys

ProviderWhere to get a key
Anthropicconsole.anthropic.com → API Keys
OpenAIplatform.openai.com → API Keys
Googleaistudio.google.com → Get API Key
OllamaNo key needed — install Ollama and pull a model