NeoHorse-1-4B needs about 4.3 GB of memory at Q4_K_M (a 2.7 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 26 quants, from 1.0 GB (Q1_0) to 4.5 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
NeoHorse-1-4B – a 4B text‑only causal language model for agent harnesses
NeoHorse-1-4B is a 4B‑parameter model derived from Qwen3.5‑4B and fine‑tuned for text‑based agent harnesses, tool use, coding, and instruction following. It supports a native context length of 262,144 tokens and is distributed only in GGUF format.
Good at
- Tool‑aware reasoning
- Agentic post‑training tasks
- Long‑context text generation
Limits
- No vision weights included
- Requires at least 2.6 GB VRAM for Q1_0 at 8 k context
- License is Apache‑2.0
Best for
- Building text‑only agent pipelines
- Coding assistance in low‑memory environments
- Long‑form content creation
Who it’s for. Developers and researchers who need a lightweight, text‑only model for agentic or tool‑using applications. The model is suitable for environments with limited GPU memory and for projects that require an Apache‑2.0 licensed LLM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q1_0 | 1.0 GB | 2.6 GB | Fits | Get file → | |
| TQ1_0 | 1.3 GB | 2.9 GB | Fits | Get file → | |
| IQ1_S | 1.4 GB | 2.9 GB | Fits | Get file → | |
| IQ1_M | 1.4 GB | 3.0 GB | Fits | Get file → | |
| TQ2_0 | 1.5 GB | 3.0 GB | Fits | Get file → | |
| IQ2_XXS | 1.5 GB | 3.1 GB | Fits | Get file → | |
| IQ2_XS | 1.6 GB | 3.2 GB | Fits | Get file → | |
| IQ2_S | 1.7 GB | 3.2 GB | Fits | Get file → | |
| IQ2_M | 1.7 GB | 3.3 GB | Fits | Get file → | |
| Q2_K_S | 1.9 GB | 3.4 GB | Fits | Get file → | |
| IQ3_XXS | 1.9 GB | 3.5 GB | Fits | Get file → | |
| Q2_K | 1.9 GB | 3.5 GB | Fits | Get file → | |
| Q3_K_S | 2.1 GB | 3.6 GB | Fits | Get file → | |
| IQ3_S | 2.1 GB | 3.7 GB | Fits | Get file → | |
| IQ3_M | 2.2 GB | 3.7 GB | Fits | Get file → | |
| Q3_K_M | 2.3 GB | 3.8 GB | Fits | Get file → | |
| Q3_K_L | 2.4 GB | 4.0 GB | Fits | Get file → | |
| IQ4_XS | 2.5 GB | 4.1 GB | Fits | Get file → | |
| Q4_K_S | 2.6 GB | 4.1 GB | Fits | Get file → | |
| IQ4_NL | 2.6 GB | 4.2 GB | Fits | Get file → | |
| Q4_K_M | 2.7 GB | 4.3 GB | Fits | Get file → | |
| Q5_K_S | 3.0 GB | 4.6 GB | Fits | Get file → | |
| Q5_K_M | 3.1 GB | 4.6 GB | Fits | Get file → | |
| Q6_K | 3.5 GB | 5.0 GB | Fits | Get file → | |
| Q6_K | 4.2 GB | 5.8 GB | Fits | Get file → | |
| Q8_0 | 4.5 GB | 6.1 GB | Fits | Get file → |
For 8 GB VRAM, Q4_K_M (2.71 GB) or Q5_K_M (3.07 GB) provide a good balance between size and fidelity. If memory is tighter, Q4_K_S (2.56 GB) or Q3_K_S (2.07 GB) are viable. For maximum fidelity, use Q8_0 (4.48 GB) or Q6_K (3.46 GB).
What GPU do you need for NeoHorse-1-4B?
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 Q3_K_L (2.4 GB file, about 4.0 GB in use).
- 6 GB VRAM: the best quant that fits is Q6_K (4.2 GB file, about 5.8 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (4.5 GB file, about 6.1 GB in use).
Which NeoHorse-1-4B quant should you choose?
Q4_K_M is the usual starting point for NeoHorse-1-4B: it balances file size and quality well. Here it is a 2.7 GB file.
The Q8_0 file is 4.5 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 1.0 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 NeoHorse-1-4B
How much VRAM does NeoHorse-1-4B need?
About 4.3 GB at Q4_K_M with an 8K context window. That is the 2.7 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 NeoHorse-1-4B on an 8 GB GPU?
Yes, with Q8_0: the 4.5 GB file needs about 6.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of NeoHorse-1-4B GGUF?
It depends on the quant: Q1_0 is 1.0 GB, TQ1_0 is 1.3 GB, IQ1_S is 1.4 GB, IQ1_M is 1.4 GB, TQ2_0 is 1.5 GB, IQ2_XXS is 1.5 GB and 20 more.
What context length does NeoHorse-1-4B 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-1-4B GGUF?
The files are on Hugging Face in the ngquocvinh/NeoHorse-1-4B-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-1-4B 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-1-4B on 8 GB VRAM?
Yes, using Q4_K_M (≈4.3 GB) or Q5_K_M (≈4.6 GB) fits within 8 GB VRAM with room for KV cache.
Does NeoHorse-1-4B support vision tasks?
No, the released GGUF files contain only text‑only weights; vision weights are not included.
What license applies to NeoHorse-1-4B?
The model is distributed under the Apache‑2.0 license.