HO

K2-Horizon-3.7B

HorizonGGUF by bartowskiUpdated 2026-10-07
Download on Hugging Face
Parameters5.06B
Context512K
Layers36
Architecturek2-horizon
Smallest quant0.0 GB
Quick answer

K2-Horizon-3.7B needs about 5.0 GB of memory at Q4_K_M (a 3.3 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 23 quants, from 0.0 GB 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: 5B‑parameter dense decoder with 512K context

K2-Horizon-3.7B is a 5B‑parameter dense decoder‑only model with a native 524,288‑token context window. It is fully open source under Apache‑2.0 and supports text input only. The model is part of the Horizon family and includes intermediate checkpoints for training analysis.

Good at

  • Strong small‑model baseline on agentic and coding benchmarks
  • 512K context for long‑form generation
  • Anti‑doom‑loop training reduces runaway generations

Limits

  • No vision or audio support
  • Requires at least 8 k context for memory estimates
  • Large memory footprint for higher‑quality quants

Best for

  • Long‑form text generation with large context
  • Agentic terminal tasks
  • Scientific reasoning benchmarks

Who it’s for. Developers and researchers who need a dense, open‑source model with a very large context window. The model is suitable for local deployment using llama.cpp or compatible tools.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 1.7 GB
Fits Get file →
IQ2_M2.0 GB 3.7 GB
Fits Get file →
Q2_K2.1 GB 3.8 GB
Fits Get file →
IQ3_XXS2.3 GB 4.0 GB
Fits Get file →
Q3_K_S2.4 GB 4.1 GB
Fits Get file →
IQ3_XS2.4 GB 4.1 GB
Fits Get file →
Q3_K_M2.5 GB 4.2 GB
Fits Get file →
Q3_K_L2.6 GB 4.3 GB
Fits Get file →
IQ3_M2.7 GB 4.4 GB
Fits Get file →
IQ4_XS3.0 GB 4.7 GB
Fits Get file →
Q4_03.1 GB 4.8 GB
Fits Get file →
Q4_K_S3.1 GB 4.8 GB
Fits Get file →
IQ4_NL3.3 GB 5.0 GB
Fits Get file →
Q4_K_M3.3 GB 5.0 GB
Fits Get file →
Q4_13.3 GB 5.0 GB
Fits Get file →
Q4_K_L3.5 GB 5.2 GB
Fits Get file →
Q5_K_S3.7 GB 5.4 GB
Fits Get file →
Q5_K_M3.9 GB 5.6 GB
Fits Get file →
Q6_K_S4.2 GB 5.9 GB
Fits Get file →
Q6_K4.4 GB 6.1 GB
Fits Get file →
Q6_K_L4.6 GB 6.3 GB
Fits Get file →
Q8_05.4 GB 7.1 GB
Fits Get file →
BF1610.1 GB 11.8 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_M (3.25 GB file, ~5 GB memory at 8 k context). With 12 GB VRAM, Q5_K_M (3.87 GB file, ~5.6 GB memory) offers better quality. For 16 GB VRAM, Q6_K (4.39 GB file, ~6.1 GB memory) provides a good balance of size and performance.

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 (2.3 GB file, about 4.0 GB in use).
  • 6 GB VRAM: the best quant that fits is Q6_K_S (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 3.3 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 the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturek2-horizon
Parameters5.06B
Layers36
Trained context524,288 tokens
KV cache (fp16)144 KB per token

Source

Quantized bybartowski
FamilyHorizon
Licenseapache-2.0
Downloads (HF)4,603
Last updated2026-10-07

Frequently asked questions about K2-Horizon-3.7B

How much VRAM does K2-Horizon-3.7B need?

About 5.0 GB at Q4_K_M with an 8K context window. That is the 3.3 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: file is 0.0 GB, IQ2_M is 2.0 GB, Q2_K is 2.1 GB, IQ3_XXS is 2.3 GB, Q3_K_S is 2.4 GB, IQ3_XS is 2.4 GB and 17 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 bartowski/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, using the Q4_K_M quantization which requires about 5 GB of VRAM at 8 k context.

Does the model support image input?

No, the model only supports text input.

What context length does the model support?

The native context window is 524,288 tokens, allowing very long generations.

Other GGUFs of K2-Horizon-3.7B

Similar models

Browse all →

More from bartowski

All by bartowski →

More Horizon models

All Horizon →

Other small models

All small models →