Skip to content

Universal Model Names

This setting gives IntenseRP the same 3 API model IDs in normal single-provider mode, no matter which provider is active. It's off by default, since it does get a bit confusing at first, but it can be a nice quality-of-life improvement if you switch providers often.

Providers in Parallel is stricter

If you have Providers in Parallel enabled, intenserp-* stays invalid because the router still needs to know which provider should handle the request.

Real-model IDs can still appear there when this setting is enabled. If an exact real-model ID would belong to more than 1 active provider, IntenseRP prefixes that conflicting ID with the provider name, like aistudio-gemini-3-1-pro-reasoner.

Settings -> API Server -> Model IDs -> Use Universal Model Names

In normal single-provider mode, GET /v1/models returns:

  • intenserp-auto
  • intenserp-reasoner
  • intenserp-chat

So if you switch or hotswap providers, you don't also have to go change the model name in SillyTavern or another client. Nice and boring, in a good way.

For providers with a real model picker, IntenseRP also lists the available real models as request-level overrides:

  • GLM Chat
  • Google AI Studio
  • QwenLM
  • Perplexity
  • HuggingChat

Those IDs are lowercase, with spaces and dots converted to -, and they keep the normal suffixes. For example, a GLM model named GLM-5.1 appears as glm-5-1-auto, glm-5-1-reasoner, and glm-5-1-chat.

In Providers in Parallel, only conflicting real-model IDs get a provider prefix. So a unique Perplexity or HuggingChat model can stay short, while an overlap like Gemini 3.1 Pro under both Google AI Studio and Perplexity becomes provider-prefixed for those providers.

intenserp-* still uses the model selected in Settings. A real-model ID switches the provider UI to that model for the request, then applies the suffix behavior on top.

Before you use it, though, there are some important things to know:

  • Provider-specific IDs like deepseek-auto or glm-chat still work even when this is enabled.
  • intenserp-reasoner is the universal version of the normal *-reasoner behavior preset.
  • Real-model IDs are only visible and accepted while Use Universal Model Names is enabled.

This replaces Better Model Names

The old experimental Better Model Names setting was removed in v2.6.3. Universal Model Names now covers that job too, without making it a separate experimental switch.

Network & API