Models
Nilux AI supports multiple AI models optimized for different tasks. This page helps you choose the right model for your work.
Available Models
Standard
The default model. Fast, capable, and cost-effective for most coding tasks.
- Best for: Everyday coding, codebase exploration, bug fixes, feature additions
- Strengths: Speed, low cost, strong code understanding
- Context window: 1M tokens
Pro
Advanced reasoning model for complex problems that require deeper analysis.
- Best for: Architecture design, complex debugging, multi-file refactors, security auditing
- Strengths: Superior reasoning, handles ambiguity, generates thorough analysis
- Context window: 1M tokens
Pricing
Prices are per 1 million tokens. You are billed for actual usage with no monthly fees. Smart caching reduces input costs by up to 98% in active sessions.
| Model | Input (per 1M) | Output (per 1M) | Cached Input (per 1M) |
|---|---|---|---|
| Standard | $0.21 | $0.42 | $0.0042 |
| Pro | $0.65 | $1.31 | $0.0054 |
See Billing & Pricing for a detailed breakdown of how billing and caching work.
Switching Models
Via Slash Command
Use the /model command inside the CLI:
/model pro
Nilux confirms the switch:
Switched to Pro
Via Configuration
Set a default model in ~/.nilux/config.json:
{
"model": "pro"
}
Or via environment variable:
export NILUX_MODEL=pro
Choosing the Right Model
| Your Task | Recommended Model |
|---|---|
| Quick code question | Standard |
| Complex debugging | Pro |
| Architecture planning | Pro |
| Daily development | Standard |
| Security auditing | Pro |
| Refactoring | Standard (simple) / Pro (complex) |
Tip: Start with Standard for most tasks. Switch to Pro when you need deeper reasoning or are working on architecture-level decisions.