# Hugging Face — Granite Switch 4.1: Chat, Decisions, and Speech from One Endpoint

- Company: Hugging Face (huggingface.co)
- Announced: 2026-10-15T20:34:01+00:00
- Category: model-update
- Coverage: not counted
- Announcement: yes
- Group: models
- Source: https://huggingface.co/blog/ibm-granite/granite-switch-4-1
- Record: https://forck.live/items/23092-granite-switch-4-1-chat-decisions-and-speech-from-one-endpoint
- Subject: Platform
- Models affected: Granite-Switch-4.1-8B, Granite-Switch-4.1-30B

IBM's Granite Switch 4.1 models (8B and 30B parameters) integrate a Decision aLoRA for general-purpose decision-making, a Guardian aLoRA for safety-related decisions, and the Granite-Speech-5.0-Turbo-CTC ASR model into a single hosted endpoint. The models reuse the base model's KV cache for adapter-based decisions, enabling fast switching between chat, decision, and speech tasks with minimal additional VRAM. The models are released with predefined adapters, and users can compose custom Granite Switch models using adapters from Granite Libraries.

## Evidence

Verbatim from https://huggingface.co/blog/ibm-granite/granite-switch-4-1:

> 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.

---

Record: https://forck.live/items/23092-granite-switch-4-1-chat-decisions-and-speech-from-one-endpoint
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md