Granite Switch 4.1 30b needs about 23.7 GB of memory at Q4_K_M (a 21.0 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 15 quants, from 14.4 GB (Q2_K) to 61.1 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.
Granite Switch 4.1 30B – 30.6B parameter LLM with modular adapters and speech-to-text
Granite Switch 4.1 30B is a 30.6B parameter LLM that composes a base model, decision and guardian adapters, and an embedded speech-to-text encoder. It supports 131072 context tokens and is released under Apache‑2.0. The model is available in multiple GGUF quantizations.
Good at
- Modular adapter composition
- Embedded speech‑to‑text
- Large context support
Limits
- No vision support
- Requires 8‑k context for memory estimates
- Apache‑2.0 license
Best for
- Chat with decision making
- Speech‑to‑text integration
- Multi‑purpose hosted endpoint
Who it’s for. Developers building a single endpoint that needs natural‑language chat, specialized decisions, and speech‑to‑text. The model is suitable for applications that require modularity and a large context window.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 14.4 GB | 17.1 GB | Too big | Get file → | |
| Q3_K_S | 15.9 GB | 18.6 GB | Too big | Get file → | |
| Q3_K_M | 17.5 GB | 20.2 GB | Too big | Get file → | |
| Q3_K_L | 18.8 GB | 21.5 GB | Too big | Get file → | |
| Q4_0 | 19.7 GB | 22.4 GB | Too big | Get file → | |
| Q4_K_S | 19.9 GB | 22.5 GB | Too big | Get file → | |
| Q4_K_M | 21.0 GB | 23.7 GB | Too big | Get file → | |
| Q4_1 | 21.5 GB | 24.1 GB | Too big | Get file → | |
| Q5_0 | 23.3 GB | 25.9 GB | Too big | Get file → | |
| Q5_K_S | 23.3 GB | 25.9 GB | Too big | Get file → | |
| Q5_K_M | 24.0 GB | 26.6 GB | Too big | Get file → | |
| Q5_1 | 25.1 GB | 27.7 GB | Too big | Get file → | |
| Q6_K | 27.1 GB | 29.7 GB | Too big | Get file → | |
| Q8_0 | 34.0 GB | 36.7 GB | Too big | Get file → | |
| BF16 | 61.1 GB (2 parts) | 63.8 GB | Too big | Get file → |
For 8‑GB VRAM, Q2_K (14.41 GB) or Q3_K_S (15.94 GB) are suitable. For 12‑GB VRAM, Q4_K_M (21.04 GB) or Q5_K_M (23.95 GB) provide a balance of size and performance. Larger models such as Q8_0 (34.05 GB) require 16‑GB or more VRAM.
What GPU do you need for Granite Switch 4.1 30b?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: the best quant that fits is Q4_K_M (21.0 GB file, about 23.7 GB in use).
- 32 GB VRAM: the best quant that fits is Q6_K (27.1 GB file, about 29.7 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (34.0 GB file, about 36.7 GB in use).
Which Granite Switch 4.1 30b quant should you choose?
Q4_K_M is the usual starting point for Granite Switch 4.1 30b: it balances file size and quality well. Here it is a 21.0 GB file.
The Q8_0 file is 34.0 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is Q2_K at 14.4 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Granite Switch 4.1 30b
How much VRAM does Granite Switch 4.1 30b need?
About 23.7 GB at Q4_K_M with an 8K context window. That is the 21.0 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Granite Switch 4.1 30b on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 17.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Granite Switch 4.1 30b GGUF?
It depends on the quant: Q2_K is 14.4 GB, Q3_K_S is 15.9 GB, Q3_K_M is 17.5 GB, Q3_K_L is 18.8 GB, Q4_0 is 19.7 GB, Q4_K_S is 19.9 GB and 9 more.
What context length does Granite Switch 4.1 30b support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.
Where can I download Granite Switch 4.1 30b GGUF?
The files are on Hugging Face in the ibm-granite/granite-switch-4.1-30b-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Granite Switch 4.1 30b use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Granite Switch 4.1 30B on 8 GB VRAM?
The smallest quantization, Q2_K, requires 17.1 GB memory at 8k context, so 8 GB VRAM is insufficient.
Does the model support vision?
No, the model does not include vision capabilities.
What license applies to this model?
The model is released under the Apache‑2.0 license.