IntegrationsNextChat

NextChat Setup

NextChat  (formerly ChatGPT-Next-Web) is a popular open-source AI client supporting Web, desktop (macOS, Windows, Linux) deployment, with a clean interface and support for multiple AI providers.

By configuring kineticRouter as your provider, you can use GPT, Claude, and other available models in NextChat with a single API Key.

NextChat supports two connection modes with kineticRouter:

ProtocolProviderEndpointSupported Models
OpenAI ChatOpenAIhttps://api.kineticrouter.comAll models (with provider/ prefix)
AnthropicAnthropichttps://api.kineticrouter.com/anthropicClaude series

Prerequisites

Setup Steps

Step 1: Open Settings

Launch NextChat and click the Settings icon in the bottom left.

This screenshot will be refreshed with kineticRouter branding during the integration documentation pass.

Step 2: Configure Model Provider and API Key

In the settings page, find the Model Provider dropdown, select the corresponding provider, and fill in the API endpoint and API Key.

Fill in the configuration based on your chosen provider:

OpenAI (supports all models)

FieldValue
Model ProviderOpenAI
API Endpointhttps://api.kineticrouter.com
API KeyYour kineticRouter API Key
Custom Model Namese.g. openai/gpt-5.3-chat,anthropic/claude-sonnet-4.6,deepseek/deepseek-v3.2

Anthropic (Claude series)

FieldValue
Model ProviderAnthropic
API Endpointhttps://api.kineticrouter.com/anthropic
API KeyYour kineticRouter API Key
Custom Model Namese.g. claude-sonnet-4.6,claude-opus-4.6,claude-haiku-4.5

This screenshot will be refreshed with kineticRouter branding during the integration documentation pass.

The Custom Model Names field supports multiple models separated by commas. These models will appear in the Model dropdown for selection.

When using the OpenAI provider, set the API Endpoint to https://api.kineticrouter.com (without /v1). NextChat appends the path automatically. Using https://api.kineticrouter.com/v1 will cause a duplicate path error.

Step 3: Select a Model

After configuration, in the Model dropdown at the bottom of the settings page, you can see the custom models you just entered.

This screenshot will be refreshed with kineticRouter branding during the integration documentation pass.

Click the dropdown to see all available providers and models.

This screenshot will be refreshed with kineticRouter branding during the integration documentation pass.

Start Using

Close settings and go back to the main screen. Select your model and start chatting.

This screenshot will be refreshed with kineticRouter branding during the integration documentation pass.

Available Model Examples

For recommended models, see the kineticRouter Model Marketplace .

Troubleshooting

Q: Error No providers support endpoint 'chat_completions'

The model name is incorrect or your account doesn’t have access to that model. Check the model name or try a different one.

Q: Error Unsupported OpenAI API endpoint

The API Endpoint is set to https://api.kineticrouter.com/v1, causing a duplicate path. Change it to https://api.kineticrouter.com (without /v1).

Q: Can the OpenAI provider call Claude and other available models?

Yes. When using the OpenAI provider, enter the full model name with provider prefix in Custom Model Names (e.g. anthropic/claude-sonnet-4.6), and kineticRouter will route to the correct provider.

Q: Does the Anthropic provider model name need a prefix?

For Anthropic, enter just claude-sonnet-4.6 without a provider prefix.