NE

NeoHorse-Jev-4B Q4_K_M unified decision + vision

NeohorseGGUF by TokenRhythmUpdated 2026-09-24
Download on Hugging Face
Parameters4.2B
Context256K
Layers32
Architectureqwen35
Smallest quant2.6 GB
Quick answer

NeoHorse-Jev-4B Q4_K_M unified decision + vision needs about 5.0 GB of memory at Q4_K_M (a 3.4 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 5 quants, from 2.6 GB (Q2_K) to 9.1 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

NeoHorse-Jev-4B Q4_K_M: 4B decision model with vision support

NeoHorse-Jev-4B is a 4.2B parameter decision model built on qwen35 architecture. It predicts application-defined decisions and probabilities using prefill-only inference. The GGUF release includes vision weights for single image and text requests.

Good at

  • Decision routing with probability outputs
  • Vision-enabled decision tasks
  • Prefill-only inference for fast response

Limits

  • No autoregressive text generation
  • Requires image+text only for vision
  • License is Apache-2.0

Best for

  • Agent workflow routing
  • Tool selection logic
  • Interactive game decision making

Who it’s for. Developers building agent systems that need fast, probability‑based decision making. The model is suitable for applications requiring image‑based decisions without full language generation.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K2.6 GB 4.2 GB
Fits Get file →
Q3_K_M2.9 GB 4.5 GB
Fits Get file →
Q4_K_M3.4 GB 5.0 GB
Fits Get file →
Q8_05.2 GB 6.7 GB
Fits Get file →
BF169.1 GB 10.7 GB
Too big Get file →

For 8 GB VRAM use Q4_K_M (3.39 GB, ~5 GB memory at 8k context). If 6 GB VRAM is available, Q3_K_M (2.95 GB, ~4.5 GB memory) is a good trade‑off. For 4 GB VRAM, Q2_K (2.6 GB, ~4.2 GB memory) can be used but may be close to limits.

What GPU do you need for NeoHorse-Jev-4B Q4_K_M unified decision + vision?

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: the best quant that fits is Q4_K_M (3.4 GB file, about 5.0 GB in use).
  • 8 GB VRAM: the best quant that fits is Q8_0 (5.2 GB file, about 6.7 GB in use).
  • 12 GB VRAM: the best quant that fits is BF16 (9.1 GB file, about 10.7 GB in use).

Which NeoHorse-Jev-4B Q4_K_M unified decision + vision quant should you choose?

Q4_K_M is the usual starting point for NeoHorse-Jev-4B Q4_K_M unified decision + vision: it balances file size and quality well. Here it is a 3.4 GB file.

The Q8_0 file is 5.2 GB, about 1.5 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 Q2_K at 2.6 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters4.2B
Layers32
Trained context262,144 tokens
KV cache (fp16)128 KB per token

Source

Quantized byTokenRhythm
FamilyNeohorse
Licenseapache-2.0
Downloads (HF)1,934
Last updated2026-09-24

Frequently asked questions about NeoHorse-Jev-4B Q4_K_M unified decision + vision

How much VRAM does NeoHorse-Jev-4B Q4_K_M unified decision + vision need?

About 5.0 GB at Q4_K_M with an 8K context window. That is the 3.4 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 NeoHorse-Jev-4B Q4_K_M unified decision + vision on an 8 GB GPU?

Yes, with Q8_0: the 5.2 GB file needs about 6.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of NeoHorse-Jev-4B Q4_K_M unified decision + vision GGUF?

It depends on the quant: Q2_K is 2.6 GB, Q3_K_M is 2.9 GB, Q4_K_M is 3.4 GB, Q8_0 is 5.2 GB, BF16 is 9.1 GB.

What context length does NeoHorse-Jev-4B Q4_K_M unified decision + vision support?

The GGUF header lists a trained context length of 262,144 tokens (256K). 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 NeoHorse-Jev-4B Q4_K_M unified decision + vision GGUF?

The files are on Hugging Face in the TokenRhythm/NeoHorse-Jev-4B-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 NeoHorse-Jev-4B Q4_K_M unified decision + vision 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 NeoHorse-Jev-4B Q4_K_M on 8 GB VRAM?

Yes. The 3.39 GB GGUF file requires about 5 GB of VRAM at 8k context, which fits within an 8 GB GPU.

Does the model support text generation?

No. It uses prefill-only inference and does not generate text.

What license applies to this model?

The model is released under the Apache-2.0 license.

Other GGUFs of NeoHorse-Jev-4B

Similar models

Browse all →

More from TokenRhythm

All by TokenRhythm →

More Neohorse models

All Neohorse →

Other small models

All small models →