DeepSeek-R1 has been making headlines for rivaling OpenAI’s O1 reasoning model while remaining fully open-source. Many users want to run it locally to ensure data privacy, reduce latency, and maintain offline access. However, fitting such a large model onto personal devices typically requires quantization (e.g. Q4_K_M), which often sacrifices accuracy (up to ~22% accuracy loss) and undermines the benefits of the local reasoning model.
We’ve solved the trade-off by quantizing the DeepSeek R1 Distilled model to 1/4 its original size—without losing any accuracy. Tests on an HP Omnibook AIPC with an AMD Ryzen™ AI 9 HX 370 processor showed a decoding speed of 66.40 tokens per second and a peak RAM usage of just 1228 MB in NexaQuant version—compared to only 25.28 tokens per second and 3788 MB RAM in the unquantized version—while NexaQuant maintaining full precision model accuracy.
NexaQuant Use Case Demo
Here’s a comparison of how a standard Q4_K_M and NexaQuant-4Bit handle a common investment banking brain teaser question. NexaQuant excels in accuracy while shrinking the model file size by 4 times.
Prompt: A Common Investment Banking BrainTeaser Question
There is a 6x8 rectangular chocolate bar made up of small 1x1 bits. We want to break it into the 48 bits. We can break one piece of chocolate horizontally or vertically, but cannot break two pieces together! What is the minimum number of breaks required?
Right Answer: 47
Example
Benchmarks
The benchmarks show that NexaQuant’s 4-bit model preserves the reasoning capacity of the original 16-bit model, delivering uncompromised performance in a significantly smaller memory & storage footprint.
Reasoning Capacity:
Example
General Capacity:
Benchmark
Full 16-bit
llama.cpp (4-bit)
NexaQuant (4-bit)
HellaSwag
35.81
34.31
34.60
MMLU
37.31
35.49
37.41
ARC Easy
67.55
54.20
65.53
MathQA
41.04
28.51
39.87
PIQA
65.56
61.70
65.07
IFEval
18.78
16.52
21.84
Run locally
NexaQuant is compatible with Nexa-SDK, Ollama, LM Studio, Llama.cpp, and any llama.cpp based project. Below, we outline multiple ways to run the model locally.
Option 1: Using Nexa SDK
Step 1: Install Nexa SDK
Follow the installation instructions in Nexa SDK's GitHub repository.
Step 2: Run the model with Nexa
Execute the following command in your terminal:
nexa run DeepSeek-R1-Distill-Qwen-1.5B-NexaQuant:q4_0
Option 2: Using llama.cpp
Step 1: Build llama.cpp on Your Device
Follow the "Building the project" instructions in the llama.cpp repository to build the project.
Step 2: Run the Model with llama.cpp
Once built, run llama-cli under <build_dir>/bin/:
bash
1./llama-cli \2 --model your/local/path/to/DeepSeek-R1-Distill-Qwen-1.5B-NexaQuant \3 --prompt 'Provide step-by-step reasoning enclosed in <think> </think> tags, followed by the final answer enclosed in \boxed{} tags.'\
In LM Studio's top panel, search for and select NexaAIDev/DeepSeek-R1-Distill-Qwen-1.5B-NexaQuant.
Click Download (if not already downloaded) and wait for the model to load.
Once loaded, go to the chat window and start a conversation.
What's next
This model is built for complex problem-solving, which is why it sometimes takes a long thinking process even for simple questions. We recognized this and are working on improving it in the next update.
Inference Nexa Quantized Deepseek-R1 distilled model on NPU.
Interested in running DeepSeek R1 on your own devices with optimized CPU, GPU, and NPU acceleration or compressing your finetuned DeepSeek-Distill-R1? Let’s chat!
