KernelLLM needs about 6.6 GB of memory at Q4_K_M (a 5.0 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 31 quants, from 2.2 GB (TQ1_0) to 16.1 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
KernelLLM – a 8B Llama‑3.1‑Instruct model with custom quantization
KernelLLM is an 8.03‑billion‑parameter Llama model derived from meta‑llama/Llama‑3.1‑8B‑Instruct. It was quantized using llama.cpp with layer bumping to improve low‑bit precision. The model supports a 131072‑token context window.
Good at
- Fine‑tuned instruction following
- Large context handling
- Custom low‑bit quantization
Limits
- No vision support
- Requires 8‑bit or higher for best performance
- License is other, not open source
Best for
- Chatbot development
- Long‑form content generation
- Research prototyping
Who it’s for. Developers and researchers who need a Llama‑3.1‑Instruct variant with improved low‑bit performance and a large context window. Suitable for local inference on GPUs with sufficient VRAM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| TQ1_0 | 2.2 GB | 3.8 GB | Fits | Get file → | |
| IQ1_S | 2.5 GB | 4.1 GB | Fits | Get file → | |
| TQ2_0 | 2.5 GB | 4.1 GB | Fits | Get file → | |
| IQ1_M | 2.7 GB | 4.3 GB | Fits | Get file → | |
| IQ2_XXS | 2.8 GB | 4.3 GB | Fits | Get file → | |
| IQ2_XS | 3.0 GB | 4.5 GB | Fits | Get file → | |
| IQ2_S | 3.1 GB | 4.6 GB | Fits | Get file → | |
| Q2_K_S | 3.1 GB | 4.7 GB | Fits | Get file → | |
| IQ2_M | 3.2 GB | 4.8 GB | Fits | Get file → | |
| Q2_K_M | 3.4 GB | 5.0 GB | Fits | Get file → | |
| IQ3_XXS | 3.4 GB | 5.0 GB | Fits | Get file → | |
| IQ3_XS | 3.6 GB | 5.2 GB | Fits | Get file → | |
| IQ3_S | 3.7 GB | 5.3 GB | Fits | Get file → | |
| Q3_K_S | 3.8 GB | 5.4 GB | Fits | Get file → | |
| IQ3_M | 3.8 GB | 5.4 GB | Fits | Get file → | |
| Q3_K_M | 4.2 GB | 5.8 GB | Fits | Get file → | |
| IQ4_XS | 4.4 GB | 6.0 GB | Fits | Get file → | |
| Q4_0 | 4.5 GB | 6.1 GB | Fits | Get file → | |
| IQ4_NL | 4.7 GB | 6.3 GB | Fits | Get file → | |
| Q4_K_S | 4.9 GB | 6.5 GB | Fits | Get file → | |
| Q4_K_M | 5.0 GB | 6.6 GB | Fits | Get file → | |
| Q4_1 | 5.0 GB | 6.6 GB | Fits | Get file → | |
| Q5_0 | 5.5 GB | 7.1 GB | Fits | Get file → | |
| Q5_K_S | 5.8 GB | 7.3 GB | Tight | Get file → | |
| Q5_K_M | 5.8 GB | 7.4 GB | Tight | Get file → | |
| Q5_1 | 6.0 GB | 7.6 GB | Tight | Get file → | |
| Q6_K_M | 6.6 GB | 8.2 GB | Too big | Get file → | |
| Q8_0 | 8.5 GB | 10.1 GB | Too big | Get file → | |
| BF16 | 11.9 GB | 13.5 GB | Too big | Get file → | |
| F16 | 11.9 GB | 13.5 GB | Too big | Get file → | |
| BF16 | 16.1 GB | 17.6 GB | Too big | Get file → |
For 8‑GB VRAM use IQ1_S (2.53 GB, 4.1 GB memory) or TQ1_0 (2.21 GB, 3.8 GB memory). For 12‑GB VRAM use Q4_0 (4.53 GB, 6.1 GB memory) or Q4_K_S (4.88 GB, 6.5 GB memory). For 16‑GB VRAM use Q5_K_M (5.82 GB, 7.4 GB memory) or Q6_K_M (6.6 GB, 8.2 GB memory).
What GPU do you need for KernelLLM?
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: the best quant that fits is TQ1_0 (2.2 GB file, about 3.8 GB in use).
- 6 GB VRAM: the best quant that fits is IQ4_XS (4.4 GB file, about 6.0 GB in use).
- 8 GB VRAM: the best quant that fits is Q5_1 (6.0 GB file, about 7.6 GB in use).
- 12 GB VRAM: the best quant that fits is Q8_0 (8.5 GB file, about 10.1 GB in use).
- 16 GB VRAM: the best quant that fits is BF16 (11.9 GB file, about 13.5 GB in use).
- 24 GB VRAM: the best quant that fits is BF16 (16.1 GB file, about 17.6 GB in use).
Which KernelLLM quant should you choose?
Q4_K_M is the usual starting point for KernelLLM: it balances file size and quality well. Here it is a 5.0 GB file.
The Q8_0 file is 8.5 GB, about 1.7 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 TQ1_0 at 2.2 GB, with the quality trade-offs that come with fewer bits. Lower bit counts mean smaller files and more quality loss.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about KernelLLM
How much VRAM does KernelLLM need?
About 6.6 GB at Q4_K_M with an 8K context window. That is the 5.0 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 KernelLLM on an 8 GB GPU?
Yes, with Q5_1: the 6.0 GB file needs about 7.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of KernelLLM GGUF?
It depends on the quant: TQ1_0 is 2.2 GB, IQ1_S is 2.5 GB, TQ2_0 is 2.5 GB, IQ1_M is 2.7 GB, IQ2_XXS is 2.8 GB, IQ2_XS is 3.0 GB and 25 more.
What context length does KernelLLM support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download KernelLLM GGUF?
The files are on Hugging Face in the Mungert/KernelLLM-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 KernelLLM use?
Hugging Face lists the license as other. 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 KernelLLM on 8 GB VRAM?
Yes, use IQ1_S or TQ1_0 quantizations which require about 4.1 GB or 3.8 GB of VRAM at 8k context.
Does KernelLLM support vision tasks?
No, the model has no vision capabilities.
What is the maximum context length of KernelLLM?
The model supports a 131072‑token context window.