AI Providers

KoreShell supports five AI providers for the AI Terminal. Configure them in Settings → AI Providers.

Anthropic (Default)

Models: Claude Haiku 3.5, Claude Sonnet 4, Claude Opus 4

Claude Haiku is the default — fast, cost-efficient, and well-suited for command explanations and shell scripting.

Setup:

1. Get an API key at console.anthropic.com
2. Settings → AI Providers → Anthropic → paste key → SAVE

Pricing: Pay-per-token. Claude Haiku is the cheapest option (~$0.25 / 1M input tokens).

OpenAI

Models: GPT-4o, GPT-4o mini, o3-mini

Setup:

1. Get an API key at platform.openai.com
2. Settings → AI Providers → OpenAI → paste key → SAVE

Google Gemini

Models: Gemini 2.0 Flash, Gemini 1.5 Pro

Setup:

1. Get an API key at aistudio.google.com
2. Settings → AI Providers → Gemini → paste key → SAVE

Gemini 2.0 Flash has a generous free tier — good for low-volume use.

OpenRouter

OpenRouter gives you access to 100+ models from a single API key.

Models: Any model listed on openrouter.ai/models

Setup:

1. Get an API key at openrouter.ai
2. Settings → AI Providers → OpenRouter → paste key → SAVE
3. In the AI tab, type the model ID directly (e.g. meta-llama/llama-3.1-70b-instruct)

Ollama (Local)

Ollama runs open-source models entirely on your machine — no internet connection or API key required. All prompts and responses stay local.

Setup:

1. Install Ollama from ollama.ai
2. Pull a model:

   ollama pull llama3.2
   ollama pull mistral
   ollama pull deepseek-r1

3. Settings → AI Providers → Ollama → set base URL (default http://localhost:11434) → SAVE
4. In the AI tab, select Ollama and choose a pulled model from the list

Recommended models for shell assistance:

Model	Pull command	Best for
Llama 3.2 3B	ollama pull llama3.2	Fast, low RAM (~2GB)
Mistral 7B	ollama pull mistral	Good balance
DeepSeek-R1 8B	ollama pull deepseek-r1	Reasoning, debugging
Qwen 2.5 Coder	ollama pull qwen2.5-coder	Code and scripts

API Key Security

All API keys are stored in the local vault (AES-256-GCM). They are sent only to the respective provider's API endpoint, never to KoreShell servers.

Switching Providers

Each conversation is locked to the provider it was started with. To use a different provider, start a New Conversation in the AI tab — the provider selector will be available.