Models
Choose a model for the agent or let Auto choose, and adjust thinking, effort, and context options.
The agent can run on models from several providers. Pick the model that suits the task, or leave the choice to Auto. The agent has the same tools whichever model you use. The model affects how it reasons and writes code.
Auto
Auto chooses a model for you and is available on every plan, including Free.
Model options
Choose a specific model in the model picker when you want it for a conversation. Models that need a paid plan show a lock. Some models offer extra options:
- Thinking: turn extended reasoning on or off.
- Effort: set how much effort the model puts into reasoning.
- Context: use a larger context window for long tasks.
Not every model accepts images. The Models settings control which models appear in the picker.
Switching models
Each conversation keeps its own model. A change applies from your next message, and the new model continues the same conversation.
Codex models
With the Codex integration, GPT models work inside CodeNeter's agent harness, using its tools to search your project, edit files, and run commands. They use your ChatGPT plan's Codex allowance at no additional CodeNeter cost.