Open-source large model of nvidia_AceReason-Nemotron-14B-GGUF - Multi-quantization versions compatible with different hardware

Nvidia AceReason Nemotron 14B GGUF

Developed by bartowski

AceReason-Nemotron-14B is a large language model with 14B parameters, offering multiple quantization versions to accommodate different hardware requirements.

Large Language Model Open Source License:Other #Multiple Quantization Options #Inference Optimization #Low RAM Adaptation

Downloads 1,772

Release Time : 5/23/2025

Model Overview

This model is a high-performance large language model suitable for various natural language processing tasks, providing versions ranging from BF16 to extremely low-bit quantization to meet the needs of different computing environments.

Model Features

Multiple Quantization Options

Offers versions ranging from BF16 to extremely low-bit quantization to adapt to different hardware environments and performance needs.

High-Quality Inference

Recommended to use quantization versions like Q6_K_L or Q5_K_M, which maintain high quality while reducing resource consumption.

Hardware Optimization

Supports online repacking for ARM and AVX machines, optimizing performance on specific hardware.

New Quantization Techniques

Utilizes novel quantization methods like I-quant to provide better performance at the same size.

Model Capabilities

Text Generation

Natural Language Understanding

Inference Task Processing

Multi-turn Dialogue

Use Cases

General Natural Language Processing

Text Generation

Generate high-quality, coherent text content

Generation quality varies depending on the quantization level

Question-Answering Systems

Build knowledge-based QA and dialogue systems

Capable of handling complex reasoning problems

Resource-Constrained Environment Applications

Mobile Device Deployment

Run low-bit quantization versions on mobile devices

Maintains usable performance under limited resources

🚀 Llamacpp imatrix Quantizations of AceReason-Nemotron-14B by nvidia

This project provides Llama.cpp imatrix quantizations of the AceReason-Nemotron-14B model by NVIDIA. It offers various quantization types to meet different performance and quality requirements.

🚀 Quick Start

Quantization Process

Use llama.cpp release b5432 for quantization.
The original model can be found at https://huggingface.co/nvidia/AceReason-Nemotron-14B.
All quantizations are made using the imatrix option with a dataset from here.

Running the Model

You can run the quantized models in LM Studio.
Or run them directly with llama.cpp, or any other llama.cpp based project.

✨ Features

Multiple Quantization Types: Offers a wide range of quantization types, including bf16, Q8_0, Q6_K_L, etc., to balance between model quality and resource usage.
Optimized for Different Hardware: Some quantization types are optimized for ARM and AVX machines, providing better performance.

📦 Installation

Downloading using huggingface-cli

Click to view download instructions

First, make sure you have hugginface-cli installed:

pip install -U "huggingface_hub[cli]"

Then, you can target the specific file you want:

huggingface-cli download bartowski/nvidia_AceReason-Nemotron-14B-GGUF --include "nvidia_AceReason-Nemotron-14B-Q4_K_M.gguf" --local-dir ./

If the model is bigger than 50GB, it will have been split into multiple files. In order to download them all to a local folder, run:

huggingface-cli download bartowski/nvidia_AceReason-Nemotron-14B-GGUF --include "nvidia_AceReason-Nemotron-14B-Q8_0/*" --local-dir ./

You can either specify a new local-dir (nvidia_AceReason-Nemotron-14B-Q8_0) or download them all in place (./)

💻 Usage Examples

Prompt format

<｜begin▁of▁sentence｜>{system_prompt}<｜User｜>{prompt}<｜Assistant｜><｜end▁of▁sentence｜><｜Assistant｜><think>

📚 Documentation

Download a file (not the whole branch) from below:

Filename	Quant type	File Size	Split	Description
AceReason-Nemotron-14B-bf16.gguf	bf16	29.55GB	false	Full BF16 weights.
AceReason-Nemotron-14B-Q8_0.gguf	Q8_0	15.70GB	false	Extremely high quality, generally unneeded but max available quant.
AceReason-Nemotron-14B-Q6_K_L.gguf	Q6_K_L	12.50GB	false	Uses Q8_0 for embed and output weights. Very high quality, near perfect, recommended.
AceReason-Nemotron-14B-Q6_K.gguf	Q6_K	12.12GB	false	Very high quality, near perfect, recommended.
AceReason-Nemotron-14B-Q5_K_L.gguf	Q5_K_L	10.99GB	false	Uses Q8_0 for embed and output weights. High quality, recommended.
AceReason-Nemotron-14B-Q5_K_M.gguf	Q5_K_M	10.51GB	false	High quality, recommended.
AceReason-Nemotron-14B-Q5_K_S.gguf	Q5_K_S	10.27GB	false	High quality, recommended.
AceReason-Nemotron-14B-Q4_K_L.gguf	Q4_K_L	9.57GB	false	Uses Q8_0 for embed and output weights. Good quality, recommended.
AceReason-Nemotron-14B-Q4_1.gguf	Q4_1	9.39GB	false	Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon.
AceReason-Nemotron-14B-Q4_K_M.gguf	Q4_K_M	8.99GB	false	Good quality, default size for most use cases, recommended.
AceReason-Nemotron-14B-Q3_K_XL.gguf	Q3_K_XL	8.61GB	false	Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability.
AceReason-Nemotron-14B-Q4_K_S.gguf	Q4_K_S	8.57GB	false	Slightly lower quality with more space savings, recommended.
AceReason-Nemotron-14B-IQ4_NL.gguf	IQ4_NL	8.55GB	false	Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference.
AceReason-Nemotron-14B-Q4_0.gguf	Q4_0	8.54GB	false	Legacy format, offers online repacking for ARM and AVX CPU inference.
AceReason-Nemotron-14B-IQ4_XS.gguf	IQ4_XS	8.12GB	false	Decent quality, smaller than Q4_K_S with similar performance, recommended.
AceReason-Nemotron-14B-Q3_K_L.gguf	Q3_K_L	7.92GB	false	Lower quality but usable, good for low RAM availability.
AceReason-Nemotron-14B-Q3_K_M.gguf	Q3_K_M	7.34GB	false	Low quality.
AceReason-Nemotron-14B-IQ3_M.gguf	IQ3_M	6.92GB	false	Medium-low quality, new method with decent performance comparable to Q3_K_M.
AceReason-Nemotron-14B-Q3_K_S.gguf	Q3_K_S	6.66GB	false	Low quality, not recommended.
AceReason-Nemotron-14B-Q2_K_L.gguf	Q2_K_L	6.53GB	false	Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable.
AceReason-Nemotron-14B-IQ3_XS.gguf	IQ3_XS	6.38GB	false	Lower quality, new method with decent performance, slightly better than Q3_K_S.
AceReason-Nemotron-14B-IQ3_XXS.gguf	IQ3_XXS	5.95GB	false	Lower quality, new method with decent performance, comparable to Q3 quants.
AceReason-Nemotron-14B-Q2_K.gguf	Q2_K	5.77GB	false	Very low quality but surprisingly usable.
AceReason-Nemotron-14B-IQ2_M.gguf	IQ2_M	5.36GB	false	Relatively low quality, uses SOTA techniques to be surprisingly usable.
AceReason-Nemotron-14B-IQ2_S.gguf	IQ2_S	5.00GB	false	Low quality, uses SOTA techniques to be usable.
AceReason-Nemotron-14B-IQ2_XS.gguf	IQ2_XS	4.70GB	false	Low quality, uses SOTA techniques to be usable.

Embed/output weights

Some of these quants (Q3_K_XL, Q4_K_L etc) are the standard quantization method with the embeddings and output weights quantized to Q8_0 instead of what they would normally default to.

ARM/AVX information

Previously, you would download Q4_0_4_4/4_8/8_8, and these would have their weights interleaved in memory in order to improve performance on ARM and AVX machines by loading up more data in one pass.

Now, however, there is something called "online repacking" for weights. details in this PR. If you use Q4_0 and your hardware would benefit from repacking weights, it will do it automatically on the fly.

As of llama.cpp build b4282 you will not be able to run the Q4_0_X_X files and will instead need to use Q4_0.

Additionally, if you want to get slightly better quality for , you can use IQ4_NL thanks to this PR which will also repack the weights for ARM, though only the 4_4 for now. The loading time may be slower but it will result in an overall speed incrase.

Click to view Q4_0_X_X information (deprecated

I'm keeping this section to show the potential theoretical uplift in performance from using the Q4_0 with online repacking.

Click to view benchmarks on an AVX2 system (EPYC7702)

model	size	params	backend	threads	test	t/s	% (vs Q4_0)
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	pp512	204.03 ± 1.03	100%
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	pp1024	282.92 ± 0.19	100%
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	pp2048	259.49 ± 0.44	100%
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	tg128	39.12 ± 0.27	100%
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	tg256	39.31 ± 0.69	100%
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	tg512	40.52 ± 0.03	100%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	pp512	301.02 ± 1.74	147%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	pp1024	287.23 ± 0.20	101%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	pp2048	262.77 ± 1.81	101%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	tg128	18.80 ± 0.99	48%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	tg256	24.46 ± 3.04	83%
qwen2 3B Q4_K_M	1.79 GiB	3.09 B	CPU	64	tg512	36.32 ± 3.59	90%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	pp512	271.71 ± 3.53	133%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	pp1024	279.86 ± 45.63	100%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	pp2048	320.77 ± 5.00	124%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	tg128	43.51 ± 0.05	111%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	tg256	43.35 ± 0.09	110%
qwen2 3B Q4_0_8_8	1.69 GiB	3.09 B	CPU	64	tg512	42.60 ± 0.31	105%

Q4_0_8_8 offers a nice bump to prompt processing and a small bump to text generation

Which file should I choose?

Click to view file selection guide

🔧 Technical Details

Quantization Method: Uses the imatrix option in llama.cpp for quantization.
Online Repacking: For ARM and AVX machines, online repacking of weights can improve performance.

📄 License

License Name: nvidia-open-model-license
License Link: https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご