Granite 4.0 H Small Heretic High needs about 24.6 GB of memory at Q4_K_XL (a 18.8 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 2 quants, from 18.8 GB (Q4_K_XL) to 64.4 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 21.5 GB.
Granite 4.0 H Small Heretic High – 32B parameter long‑context model
Granite 4.0 H Small Heretic High is a 32.2B parameter model with a 1,048,576 token context window. It is a mixture of experts fine‑tuned for instruction following and tool‑calling. The model is released under Apache‑2.0 and supports multiple languages.
Good at
- Tool‑calling
- Long‑context instruction following
- Enterprise‑grade safety
Limits
- No vision support
- Requires large VRAM for full precision
- License allows commercial use but requires attribution
Best for
- Building AI assistants
- Enterprise chatbots
- Tool‑integrated workflows
Who it’s for. Developers and data scientists building AI assistants or enterprise chatbots that need robust instruction following and tool‑calling. The model is suitable for teams that can allocate sufficient GPU memory and comply with the Apache‑2.0 license.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_XL | 18.8 GB | 24.6 GB | Too big | Get file → | |
| BF16 | 64.4 GB (2 parts) | 70.3 GB | Too big | Get file → |
For 8 GB VRAM use the Q4_K_XL quantized file (18.76 GB, ~24.6 GB memory at 8 k context). For higher precision use the BF16 shard (64.45 GB, ~70.3 GB memory at 8 k context). The Q4_K_XL is the most memory‑efficient option.
What GPU do you need for Granite 4.0 H Small Heretic High?
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: no quant fits at an 8K context.
- 32 GB VRAM: the best quant that fits is Q4_K_XL (18.8 GB file, about 24.6 GB in use).
- 48 GB VRAM: the best quant that fits is Q4_K_XL (18.8 GB file, about 24.6 GB in use).
Which Granite 4.0 H Small Heretic High quant should you choose?
Q4_K_XL (18.8 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Granite 4.0 H Small Heretic High
How much VRAM does Granite 4.0 H Small Heretic High need?
About 24.6 GB at Q4_K_XL with an 8K context window. That is the 18.8 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 4.0 H Small Heretic High on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 24.6 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Granite 4.0 H Small Heretic High GGUF?
It depends on the quant: Q4_K_XL is 18.8 GB, BF16 is 64.4 GB.
What context length does Granite 4.0 H Small Heretic High support?
The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 5.4 GB at 8K and 21.5 GB at 32K.
Where can I download Granite 4.0 H Small Heretic High GGUF?
The files are on Hugging Face in the rodrigomt/granite-4.0-h-small-heretic-high-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 4.0 H Small Heretic High 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 this model on 8 GB VRAM?
Yes, using the Q4_K_XL quantized file requires about 24.6 GB memory at 8 k context, which is within 8 GB VRAM limits.
Does this 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.