Pricing
Free to start, powerful when you need it
Code for free with your own models. Upgrade to a paid plan for premium models managed by Navin, with no API keys to configure.
Navin Free
Code for free with your own models.
- Windows, macOS and Linux apps
- Your API keys: OpenAI, Anthropic, Gemini, Kimi, MiniMax, GLM, OpenRouter...
- Local models: Ollama, LM Studio, llama.cpp
- Local search: grep, Git, LSP, Tree-sitter
- navin.live account and basic sync
- 2 devices
Navin Plus
Premium models included for regular use.
6M equivalent tokens / month
- Premium models managed by Navin
- Latest MiniMax, GLM and Kimi
- Automatic routing to the right model
- 1 active agent, 2 devices
- Cloud history and full sync
- Free or local fallback, no interruption
Navin Pro
Powerful agents for professional developers.
22M equivalent tokens / month
- All premium models, automatic selection
- 3 concurrent agents, 5 devices
- Long agentic tasks, extended context
- Expert models used more frequently
- Higher priority
- Full sync
Navin Ultra
Intensive use and complex projects.
60M equivalent tokens / month
- 8 concurrent agents, 10 devices
- Large codebases, maximum context
- Expert models on every complex task
- Long autonomous tasks
- Maximum priority
- Advanced history and sync
Monthly AI credits are debited based on the real cost of the model used (Navin equivalent tokens). When the quota is reached, Navin automatically falls back to your keys, free models or a local model - it never stops.
You can always bring your own models
On every plan, including Free, you can use your own API keys and local models. They stay stored on your machine (Credential Manager, Keychain, Secret Service) and never pass through our servers.
Frequently asked questions
What does the free plan include?+
Navin Free is a real product, not a trial: Windows, macOS and Linux apps, a navin.live account, your own API keys, local and free models, local code search, terminal, skills and tools. No credit card required.
What are monthly AI credits?+
Paid plans include a monthly quota of Navin equivalent tokens (6M, 22M or 60M). Every call is debited based on the real cost of the model used: a task on an economical model consumes far less than one on an expert model.
What happens when my quota is reached?+
Navin never stops abruptly. As you approach the limit it favors economical models; at 100% it continues with free or local models, your own API keys, or extra credits.
Which models are included in paid plans?+
The latest models via OpenRouter: Kimi K3, GLM 5.2, MiniMax M3 (the default model), Claude, GPT, Gemini, Qwen, Grok and more. The Navin router automatically picks the right model for each step to optimize quality and budget.
Can I cancel anytime?+
Yes, no commitment. You keep the managed models until the end of the paid period, then your account automatically returns to Navin Free with your conversations and settings intact.