Instructions to use eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M

Ollama
How to use eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF with Ollama:
```
ollama run hf.co/eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF with Docker Model Runner:
```
docker model run hf.co/eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M
```

Lemonade

How to use eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull eaddario/DeepSeek-R1-Distill-Llama-8B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.DeepSeek-R1-Distill-Llama-8B-GGUF-Q4_K_M

List all available models

lemonade list

Atomic Chat

DeepSeek-R1-Distill-Llama-8B-GGUF / scores /DeepSeek-R1-Distill-Llama-8B-iq4_nl.ppx

eaddario

Generate Perplexity, KLD, ARC, HellaSwag, MMLU, Truthful QA and WinoGrande scores

6fcfa91 verified over 1 year ago

Raw

History Blame

1.07 kB

	====== Perplexity statistics ======
	Mean PPL(Q) : 14.243482 ± 0.113690
	Mean PPL(base) : 13.052024 ± 0.108483
	Cor(ln(PPL(Q)), ln(PPL(base))): 96.12%
	Mean ln(PPL(Q)/PPL(base)) : 0.087356 ± 0.002292
	Mean PPL(Q)/PPL(base) : 1.091285 ± 0.002502
	Mean PPL(Q)-PPL(base) : 1.191458 ± 0.031368

	====== KL divergence statistics ======
	Mean KLD: 0.240811 ± 0.001175
	Maximum KLD: 14.633161
	99.9% KLD: 4.552272
	99.0% KLD: 2.206026
	99.0% KLD: 2.206026
	Median KLD: 0.107760
	10.0% KLD: 0.005211
	5.0% KLD: 0.001413
	1.0% KLD: 0.000154
	Minimum KLD: -0.000001

	====== Token probability statistics ======
	Mean Δp: -4.129 ± 0.033 %
	Maximum Δp: 97.466%
	99.9% Δp: 52.946%
	99.0% Δp: 23.458%
	95.0% Δp: 8.259%
	90.0% Δp: 3.367%
	75.0% Δp: 0.105%
	Median Δp: -0.402%
	25.0% Δp: -5.853%
	10.0% Δp: -17.104%
	5.0% Δp: -27.264%
	1.0% Δp: -55.844%
	0.1% Δp: -87.698%
	Minimum Δp: -99.895%
	RMS Δp : 13.331 ± 0.064 %
	Same top p: 79.968 ± 0.105 %