K2 Horizon 3.7B needs about 4.9 GB of memory at Q4_K_M (a 3.2 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 30 quants, from 1.2 GB (Q1_0) to 10.1 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.8 GB.
K2 Horizon 3.7B – a dense decoder-only model with 512K context
K2 Horizon 3.7B is a 5.1B‑parameter dense model with a native 524,288‑token context window. It is fully open source, released under Apache‑2.0, and quantized into multiple GGUF formats for different memory budgets.
Good at
- Large context generation
- Stable agentic tasks
- Reasoning benchmarks
Limits
- Requires MBZUAI‑IFM llama.cpp fork
- No vision support
- Large VRAM needed for higher quants
Best for
- Long‑form text generation
- Agentic workflow scripts
- Research on context‑sensitive models
Who it’s for. Developers and researchers who need a high‑capacity, open‑source language model with a very large context window. The model is suitable for local deployment on systems with sufficient GPU memory.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q1_0 | 1.2 GB | 2.9 GB | Fits | Get file → | |
| IQ1_S | 1.5 GB | 3.2 GB | Fits | Get file → | |
| IQ1_M | 1.5 GB | 3.3 GB | Fits | Get file → | |
| IQ2_XXS | 1.7 GB | 3.4 GB | Fits | Get file → | |
| IQ2_XS | 1.8 GB | 3.5 GB | Fits | Get file → | |
| IQ2_S | 1.9 GB | 3.6 GB | Fits | Get file → | |
| Q2_0 | 2.0 GB | 3.7 GB | Fits | Get file → | |
| IQ2_M | 2.0 GB | 3.7 GB | Fits | Get file → | |
| Q2_K_S | 2.0 GB | 3.7 GB | Fits | Get file → | |
| Q2_K | 2.1 GB | 3.9 GB | Fits | Get file → | |
| IQ3_XXS | 2.2 GB | 3.9 GB | Fits | Get file → | |
| IQ3_XS | 2.4 GB | 4.1 GB | Fits | Get file → | |
| Q3_K_S | 2.4 GB | 4.1 GB | Fits | Get file → | |
| IQ3_S | 2.4 GB | 4.2 GB | Fits | Get file → | |
| IQ3_M | 2.5 GB | 4.2 GB | Fits | Get file → | |
| Q3_K_M | 2.6 GB | 4.3 GB | Fits | Get file → | |
| Q3_K_L | 2.8 GB | 4.5 GB | Fits | Get file → | |
| IQ4_XS | 2.9 GB | 4.6 GB | Fits | Get file → | |
| Q4_0 | 3.0 GB | 4.7 GB | Fits | Get file → | |
| IQ4_NL | 3.0 GB | 4.7 GB | Fits | Get file → | |
| Q4_K_S | 3.0 GB | 4.7 GB | Fits | Get file → | |
| Q4_K_M | 3.2 GB | 4.9 GB | Fits | Get file → | |
| Q4_1 | 3.3 GB | 5.0 GB | Fits | Get file → | |
| Q5_0 | 3.6 GB | 5.3 GB | Fits | Get file → | |
| Q5_K_S | 3.6 GB | 5.3 GB | Fits | Get file → | |
| Q5_K_M | 3.6 GB | 5.4 GB | Fits | Get file → | |
| Q5_1 | 3.9 GB | 5.6 GB | Fits | Get file → | |
| Q6_K | 4.2 GB | 5.9 GB | Fits | Get file → | |
| Q8_0 | 5.4 GB | 7.1 GB | Fits | Get file → | |
| BF16 | 10.1 GB | 11.8 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_0 (3.02 GB, ~4.7 GB memory at 8k context). For 12 GB VRAM, Q5_0 (3.57 GB, ~5.3 GB memory) is a good balance. If you have 16 GB or more, consider Q6_K (4.16 GB, ~5.9 GB memory) for higher quality.
What GPU do you need for K2 Horizon 3.7B?
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 IQ3_XXS (2.2 GB file, about 3.9 GB in use).
- 6 GB VRAM: the best quant that fits is Q6_K (4.2 GB file, about 5.9 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (5.4 GB file, about 7.1 GB in use).
- 12 GB VRAM: the best quant that fits is BF16 (10.1 GB file, about 11.8 GB in use).
Which K2 Horizon 3.7B quant should you choose?
Q4_K_M is the usual starting point for K2 Horizon 3.7B: it balances file size and quality well. Here it is a 3.2 GB file.
The Q8_0 file is 5.4 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 Q1_0 at 1.2 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 K2 Horizon 3.7B
How much VRAM does K2 Horizon 3.7B need?
About 4.9 GB at Q4_K_M with an 8K context window. That is the 3.2 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 K2 Horizon 3.7B on an 8 GB GPU?
Yes, with Q8_0: the 5.4 GB file needs about 7.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of K2 Horizon 3.7B GGUF?
It depends on the quant: Q1_0 is 1.2 GB, IQ1_S is 1.5 GB, IQ1_M is 1.5 GB, IQ2_XXS is 1.7 GB, IQ2_XS is 1.8 GB, IQ2_S is 1.9 GB and 24 more.
What context length does K2 Horizon 3.7B support?
The GGUF header lists a trained context length of 524,288 tokens (512K). The KV cache grows with the context you use: about 1.2 GB at 8K and 4.8 GB at 32K.
Where can I download K2 Horizon 3.7B GGUF?
The files are on Hugging Face in the NANI-Nithin/K2-Horizon-3.7B-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 K2 Horizon 3.7B 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 K2 Horizon 3.7B on 8 GB VRAM?
Yes, the Q4_0 quantization requires about 4.7 GB of VRAM at 8k context, which fits within an 8 GB GPU.
Does this model support vision or image inputs?
No, K2 Horizon 3.7B is a text‑only decoder‑only model and does not include vision capabilities.
What fork of llama.cpp is needed to run these GGUFs?
The MBZUAI‑IFM llama.cpp fork, branch `model/K2Horizon`, is required to load the k2‑horizon architecture.