Skip to main content
Open the Coreply app. The main screen shows all settings described below. Settings are saved automatically as you change them.

​
Connection

These three fields tell Coreply which model to call and how to authenticate.
​
Base URL
string
default:"https://api.openai.com/v1/"
required
The base URL of an OpenAI-compatible API endpoint. Change this when using a provider other than OpenAI — for example, https://generativelanguage.googleapis.com/v1beta/openai/ for Google AI Studio or https://api.groq.com/openai/v1/ for Groq.
​
API Key
string
required
Your authentication key from the AI provider. The field is masked by default — tap the eye icon to reveal it.
​
Model Name
string
default:"gpt-4.1-mini"
required
The model identifier to pass in API requests (for example, gpt-4.1-mini, gemini-2.0-flash, llama-3.3-70b-versatile). The exact value depends on your provider — see AI Providers for recommended models.

​
Simple vs advanced mode

The Simple and Advanced radio buttons switch between two ways of configuring the request body sent to the API.

​
Simple mode

Simple mode exposes individual fields that Coreply uses to build a standard chat completion request automatically.
​
System Prompt
string
Instructions prepended to every API request as a system message. Edit this to change the assistant’s persona, output format, or language style. A default system prompt is pre-filled.
​
Temperature
number
default:"0.3"
Controls randomness in the model’s output. Range: 0 to 1. Lower values produce more predictable suggestions; higher values produce more varied ones.
​
Top-P
number
default:"1.0"
Controls diversity of token selection via nucleus sampling. Range: 0 to 1. The default of 1.0 lets temperature alone govern randomness.

​
Advanced mode

Advanced mode replaces the individual fields with a free-form Request Body text area that accepts a full JSON template with Mustache variables. Use it when you need to control parameters that simple mode doesn’t expose — such as max_tokens, non-standard model parameters, or a completely different prompt structure. See the Advanced Config page for a full walkthrough and the default template.

​
Suggestion display

​
Suggestion mode
string
default:"Bubble and inline"
Controls where suggestions appear on screen. Options:
  • Bubble below text field only — a floating bubble appears just below the text input.
  • Inline only — the suggestion appears as ghost text inside the text field.
  • Bubble and inline — both are shown simultaneously (default).

​
Customizations

​
Debounce (ms)
integer
default:"350"
How long (in milliseconds) Coreply waits after you stop typing before sending a request to the AI provider. Range: 150–1200ms. Lower values give faster suggestions at the cost of more API calls; higher values reduce API calls but add a delay.
​
Enable Regex Filter
boolean
default:"false"
When enabled, Coreply only requests a suggestion when the text you have typed matches the regex pattern below.
​
Typing Regex Pattern
string
default:"^.*[\\s.!?,;:]$"
The regex pattern used to decide when to trigger a suggestion. Only evaluated when Enable Regex Filter is on. For example, the default pattern triggers suggestions after a word ending in a space or punctuation mark.

​
Other settings

​
Show errors
boolean
default:"true"
When enabled, API errors and network failures are shown in the on-screen overlay so you can see what went wrong in real time.
​
Select Apps
list
The apps in which Coreply is active. Tap Select Apps to open a list of supported apps installed on your device and toggle them individually. All supported apps are enabled by default.