HO

K2 Horizon MoVA 36B A4B ROCmFP4

HorizonGGUF by kingjones777Updated 2026-09-06
Download on Hugging Face
Parameters37.4B
Context512K
Layers48
Architecturek2-horizon
Smallest quant19.9 GB
Quick answer

K2 Horizon MoVA 36B A4B ROCmFP4 needs about 22.2 GB of memory at Q4_0 (a 20.0 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 4 quants, from 19.9 GB (Q4_0) to 39.1 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 6.4 GB.

K2 Horizon MoVA 36B A4B ROCmFP4 – a 36B Mixture‑of‑Experts reasoning model

K2 Horizon MoVA 36B A4B is a 36.4‑billion‑parameter MoE model with 4.0‑billion active parameters. It supports a 524,288‑token context window and is optimized for reasoning tasks. The model is released under the Apache‑2.0 license.

Good at

  • Reasoning and chain‑of‑thought generation
  • Large context handling
  • Mixture‑of‑Experts efficiency

Limits

  • No vision support
  • Requires ROCmFP4/ROCmFPX build
  • Large VRAM needed for full precision

Best for

  • Long‑form reasoning
  • Complex problem solving
  • Agentic tool‑use scenarios

Who it’s for. Developers building reasoning‑heavy applications on AMD hardware. Researchers needing a high‑capacity MoE model for long‑context tasks.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_019.9 GB 22.1 GB
Too big Get file →
Q4_020.0 GB 22.2 GB
Too big Get file →
Q4_021.3 GB 23.4 GB
Too big Get file →
Q8_039.1 GB 41.3 GB
Too big Get file →

For 8 GB VRAM, use the Q4_0 quantised file (19.94 GB, ~22.1 GB memory at 8 k context). For 16 GB VRAM, the larger Q4_0 file (21.28 GB, ~23.4 GB memory) is suitable. For 32 GB VRAM, the Q8_0 file (39.15 GB, ~41.3 GB memory) offers the highest precision.

What GPU do you need for K2 Horizon MoVA 36B A4B ROCmFP4?

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: the best quant that fits is Q4_0 (21.3 GB file, about 23.4 GB in use).
  • 32 GB VRAM: the best quant that fits is Q4_0 (21.3 GB file, about 23.4 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (39.1 GB file, about 41.3 GB in use).

Which K2 Horizon MoVA 36B A4B ROCmFP4 quant should you choose?

Q4_0 (20.0 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.

If memory is tight, the smallest file is Q4_0 at 19.9 GB, with the quality trade-offs that come with fewer bits. 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

Architecturek2-horizon
Parameters37.4B (approximate, mixture of experts)
Active parameters4B
Layers48
Trained context524,288 tokens
KV cache (fp16)192 KB per token

Source

Quantized bykingjones777
FamilyHorizon
Licenseapache-2.0
Downloads (HF)3,631
Last updated2026-09-06

Frequently asked questions about K2 Horizon MoVA 36B A4B ROCmFP4

How much VRAM does K2 Horizon MoVA 36B A4B ROCmFP4 need?

About 22.2 GB at Q4_0 with an 8K context window. That is the 20.0 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 K2 Horizon MoVA 36B A4B ROCmFP4 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 22.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of K2 Horizon MoVA 36B A4B ROCmFP4 GGUF?

It depends on the quant: Q4_0 is 19.9 GB, Q4_0 is 20.0 GB, Q4_0 is 21.3 GB, Q8_0 is 39.1 GB.

What context length does K2 Horizon MoVA 36B A4B ROCmFP4 support?

The GGUF header lists a trained context length of 524,288 tokens (512K). The KV cache grows with the context you use: about 1.6 GB at 8K and 6.4 GB at 32K.

Where can I download K2 Horizon MoVA 36B A4B ROCmFP4 GGUF?

The files are on Hugging Face in the kingjones777/K2-Horizon-MoVA-36B-A4B-ROCmFP4-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 MoVA 36B A4B ROCmFP4 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?

Use the Q4_0 quantised file; it requires approximately 22.1 GB memory at 8 k context, so 8 GB VRAM is insufficient.

Does this model support vision inputs?

No, the model has no vision capabilities.

What license applies to this model?

The model is released under the Apache‑2.0 license.

Other GGUFs of K2-Horizon-MoVA-36B-A4B

Similar models

Browse all →

More from kingjones777

All by kingjones777 →

More Horizon models

All Horizon →

Other large models

All large models →