Instructions to use anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6")
model = AutoModelForCausalLM.from_pretrained("anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6

SGLang

How to use anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6 with Docker Model Runner:
```
docker model run hf.co/anthonyw448/DeepSeek-R1-Distill-Llama-70B_EXL2_5.5bpw_h6
```

Configuration Parsing Warning:In config.json: "quantization_config.bits" must be an integer

DeepSeek-R1-Distill-Llama-70B EXL2 5.5bpw h6

EXL2 quantization of deepseek-ai/DeepSeek-R1-Distill-Llama-70B

Quantization Details

Format: EXL2
Bits per weight: 5.5bpw
Head bits: 6
Quantized with: ExLlamaV2 0.3.1
Calibration: Default ExLlamaV2 calibration dataset

Hardware Requirements

Minimum VRAM: ~40GB (e.g. 2x RTX 3090)
Recommended: 48GB+ for comfortable context

Usage

Load with TabbyAPI or text-generation-webui with ExLlamaV2 loader.

Original Model

See deepseek-ai/DeepSeek-R1-Distill-Llama-70B for full details, license, and usage recommendations.

Downloads last month: 34