HO

K2 Horizon 0.9B

HorizonGGUF by NANI-NithinUpdated 2026-09-03
Download on Hugging Face
Parameters1.08B
Context128K
Layers28
Architecturek2-horizon
Smallest quant0.2 GB
Quick answer

K2 Horizon 0.9B needs about 1.6 GB of memory at Q4_K_M (a 0.7 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 30 quants, from 0.2 GB (Q1_0) to 2.2 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.9 GB.

K2 Horizon 0.9B – a 0.9B‑parameter decoder‑only model with 128K context

K2 Horizon 0.9B is a compact dense model in the Horizon family. It supports up to 131,072 tokens with YaRN RoPE scaling. The model is fully open and trained with multi‑teacher distillation for math, code, STEM, and instruction following.

Good at

  • Reasoning tasks
  • Math and coding benchmarks
  • Instruction following

Limits

  • Requires MBZUAI‑IFM llama.cpp fork
  • No vision support
  • Large context window may need more VRAM

Best for

  • Local inference on 8‑GB VRAM
  • Educational coding assistants
  • Math problem solving

Who it’s for. Developers and researchers who need a small, open‑source model with a large context window for reasoning tasks. The model is suitable for local deployment on consumer GPUs.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q1_00.2 GB 1.2 GB
Fits Get file →
IQ1_S0.3 GB 1.3 GB
Fits Get file →
IQ1_M0.3 GB 1.3 GB
Fits Get file →
IQ2_XXS0.3 GB 1.3 GB
Fits Get file →
IQ2_XS0.4 GB 1.3 GB
Fits Get file →
Q2_00.4 GB 1.4 GB
Fits Get file →
IQ2_S0.4 GB 1.4 GB
Fits Get file →
IQ2_M0.4 GB 1.4 GB
Fits Get file →
Q2_K_S0.4 GB 1.4 GB
Fits Get file →
Q2_K0.4 GB 1.4 GB
Fits Get file →
IQ3_XXS0.5 GB 1.4 GB
Fits Get file →
IQ3_XS0.5 GB 1.5 GB
Fits Get file →
Q3_K_S0.5 GB 1.5 GB
Fits Get file →
IQ3_S0.5 GB 1.5 GB
Fits Get file →
IQ3_M0.5 GB 1.5 GB
Fits Get file →
Q3_K_M0.5 GB 1.5 GB
Fits Get file →
Q3_K_L0.6 GB 1.6 GB
Fits Get file →
IQ4_XS0.6 GB 1.6 GB
Fits Get file →
Q4_00.6 GB 1.6 GB
Fits Get file →
IQ4_NL0.6 GB 1.6 GB
Fits Get file →
Q4_K_S0.6 GB 1.6 GB
Fits Get file →
Q4_K_M0.7 GB 1.6 GB
Fits Get file →
Q4_10.7 GB 1.7 GB
Fits Get file →
Q5_00.8 GB 1.7 GB
Fits Get file →
Q5_K_S0.8 GB 1.7 GB
Fits Get file →
Q5_K_M0.8 GB 1.7 GB
Fits Get file →
Q5_10.8 GB 1.8 GB
Fits Get file →
Q6_K0.9 GB 1.9 GB
Fits Get file →
Q8_01.1 GB 2.1 GB
Fits Get file →
BF162.2 GB 3.1 GB
Fits Get file →

For 8‑GB VRAM, use Q4_0 (0.63 GB) or Q4_K_M (0.67 GB). If 12‑GB VRAM is available, Q5_0 (0.76 GB) or Q5_K_M (0.77 GB) provide better quality. Larger quants such as Q8_0 (1.15 GB) or BF16 (2.16 GB) require more memory but offer higher precision.

What GPU do you need for K2 Horizon 0.9B?

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 BF16 (2.2 GB file, about 3.1 GB in use).

Which K2 Horizon 0.9B quant should you choose?

Q4_K_M is the usual starting point for K2 Horizon 0.9B: it balances file size and quality well. Here it is a 0.7 GB file.

The Q8_0 file is 1.1 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 0.2 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

Architecturek2-horizon
Parameters1.08B
Layers28
Trained context131,072 tokens
KV cache (fp16)56 KB per token

Source

Quantized byNANI-Nithin
FamilyHorizon
Licenseapache-2.0
Downloads (HF)2,386
Last updated2026-09-03

Frequently asked questions about K2 Horizon 0.9B

How much VRAM does K2 Horizon 0.9B need?

About 1.6 GB at Q4_K_M with an 8K context window. That is the 0.7 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 0.9B on an 8 GB GPU?

Yes, with BF16: the 2.2 GB file needs about 3.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of K2 Horizon 0.9B GGUF?

It depends on the quant: Q1_0 is 0.2 GB, IQ1_S is 0.3 GB, IQ1_M is 0.3 GB, IQ2_XXS is 0.3 GB, IQ2_XS is 0.4 GB, Q2_0 is 0.4 GB and 24 more.

What context length does K2 Horizon 0.9B support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.5 GB at 8K and 1.9 GB at 32K.

Where can I download K2 Horizon 0.9B GGUF?

The files are on Hugging Face in the NANI-Nithin/K2-Horizon-0.9B-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 0.9B 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 0.9B on 8 GB VRAM?

Yes, using Q4_0 or Q4_K_M quantizations will fit within 8 GB VRAM with an estimated 1.6 GB memory usage at 8k context.

Does this model support vision?

No, K2 Horizon 0.9B is a text‑only decoder‑only model.

Which fork of llama.cpp is required?

The MBZUAI‑IFM llama.cpp fork, branch model/K2Horizon, is required to run the k2‑horizon architecture.

Other GGUFs of K2-Horizon-0.9B

Similar models

Browse all →

More from NANI-Nithin

All by NANI-Nithin →

More Horizon models

All Horizon →

Other tiny models

All tiny models →