AI Providers & Models

KnightCode is built on a bring-your-own-key (BYOK) architecture. It does not run proprietary proxy servers or charge markup fees for inference. Instead, KnightCode exclusively uses OpenRouter as the gateway to connect to language models.

This means you only need to obtain and configure a single API key from OpenRouter to access a wide variety of state-of-the-art models.

Managing Models and Keys

To view available models or switch between them:

1. Type /models in the TUI input bar to open the Select Model dialog.
2. Search and select from the list of supported models.

The default model on a fresh install is Nemotron 3 Ultra 550B (free) (nvidia/nemotron-3-ultra-550b-a55b:free).

To re-run key configurations:

1. Type /setup in the input bar.
2. Follow the onboarding prompt to input your OpenRouter API key.

Secure Storage

API keys and search credentials are stored locally in the configuration directory:

• Path: ~/.knightcode/credentials.json (on macOS/Linux) or %USERPROFILE%\.knightcode\credentials.json (on Windows). If the KNIGHTCODE_HOME environment variable is set, the credentials will be stored at $KNIGHTCODE_HOME/credentials.json instead.

On macOS, Linux, and other Unix-like systems, KnightCode configures the file permissions to 0600 (owner-only read and write) to prevent other users or unprivileged processes on the system from reading your API keys.

To completely wipe all keys, you can delete the credentials.json file. Note that this does not clear keys provided via environment variables (such as OPENROUTER_API_KEY); you will need to unset those environment variables separately.

Onboarding Shortlist

The onboarding wizard offers a curated shortlist (free → frontier) to get you started quickly:

All Supported Models

The /models dialog lists the full set of supported models. All run through OpenRouter, and all support reasoning/thinking effort.

• Free:
  • nvidia/nemotron-3-ultra-550b-a55b:free (Nemotron 3 Ultra 550B)
  • openai/gpt-oss-120b:free (GPT-OSS 120B)
  • nex-agi/nex-n2-pro:free (Nex N2 Pro)
• Standard & Frontier:
  • anthropic/claude-opus-4.8 (Claude Opus 4.8)
  • anthropic/claude-sonnet-4.6 (Claude Sonnet 4.6)
  • anthropic/claude-haiku-4.5 (Claude Haiku 4.5)
  • openai/gpt-5.5-pro (GPT-5.5 Pro)
  • openai/gpt-5.5 (GPT-5.5)
  • openai/gpt-5.4-mini (GPT-5.4 Mini)
  • openai/gpt-5.3-codex (GPT-5.3 Codex)
  • qwen/qwen3.7-plus (Qwen3.7 Plus)
  • z-ai/glm-5.2 (GLM 5.2)
  • moonshotai/kimi-k2.7-code (Kimi K2.7 Code)
  • minimax/minimax-m2.7 (MiniMax M2.7)