Granite Switch 4.1: Chat, Decisions, and Speech from One Endpoint

Introduced last spring, the Granite Switch project enables a modular approach to model development. Now Granite Switch 4.1 models, armed with a Decision adapter and the fastest ASR model on the OpenASR leaderboard, can move efficiently between natural-language responses, general and specialized decisions, and speech-to-text tasks—all from one hosted endpoint.

Specifically, Granite-Switch-4.1-8B and -30B models are configured to include:

All three of these technologies (along with the underlying base model) are composed into one model, taking up approximately the same amount of VRAM on a GPU as the base model would by itself. That means more capabilities with fewer GPUs.

Moving forward, we see tremendous opportunity for the community to compose and extend Granite Switch models with custom and third party adapters. We plan to release an Adapter SDK for training adapters in Mellea , along side other tutorials and Granite Switch support. These adapters could be specialized classification adapters to augment decision-based workflows or adapters for specialized language tasks. You could even use custom adapters to teach a 3B parameter version of Granite to play the classic game Doom.

Similarly to Jev, our custom 3B parameter Granite Switch model made split second decisions to navigate the Doom game, but was able to do so at 60 - 100 decisions / second, compared to Jev’s reported 10 decisions / second. 1 Unlike Jev, this Granite Switch model can also narrate and respond to your questions while it is playing.

In a “Race” between Granite-Switch-4.1-30B (frozen base Granite + 4 specialized aLoRAs) and Jev-27B (frozen base Qwen model + 1 generic decision LoRA), Granite-Switch-4.1-30B was able to complete a real world, 5-step RAG workflow much faster than its competitor.

Coming in 8b and 30b sizes, Granite Switch 4.1 can fit wherever your workload runs. The model runs in vllm with a plug-in (see Granite Switch repo below), and has native inference support in llama.cpp and downstream platforms which allows the model to seamlessly swap in for /v1/systemone API calls anywhere they're used, while also servicing traditional chat calls, all with a shared prefix cache.

And because the Switch adapters were built as cache-efficient Activated LoRAs , the KV cache created from a base model generation can be reused for any subsequent adapter-based decisions. Combined with our team's recent vLLM contributions to speed up low rank adapter inferencing, as well as a co-located ASR encoder, Granite Switch 4.1 models can switch between a chat request and a decision request and a spoken request, with blazing fast speeds.

Check out our Speech-to-Decisions HF space where granite-switch-4.1-8b transcribes audio, generates summaries, screens for harmful content, and tests 15 other user-defined flags—all running on a single endpoint in real time.

Decision models aren’t new. Our previous release of Granite Libraries features a repository of specialized “decision model” adapters and other specialized task adapters that can be composed into custom Granite Switch models. And while Jev and other closed decision models focus on generality, we’ve found that many of our specialized granitelib adapters can outperform generalists on specialized classification tasks.