Model Overview

• Model Optimizations:
  • Weight quantization: FP8
  • Activation quantization: FP8
• Release Date: 1/29/2025

Quantized version of deepseek-ai/DeepSeek-R1-0528-Qwen3-8B to FP8 data type, ready for inference with SGLang >= 0.3 or vLLM >= 0.5.2. This optimization reduces the number of bits per parameter from 16 to 8, reducing the disk size and GPU memory requirements by approximately 50%. Only the weights and activations of the linear operators within transformers blocks are quantized.

Deployment

Use with SGLang

python -m sglang.launch_server --model-path JamAndTeaStudios/DeepSeek-R1-0528-Qwen3-8B-FP8-Dynamic \
--port 30000 --host 0.0.0.0

Use with vLLM

python -m vllm.entrypoints.openai.api_server --model JamAndTeaStudios/DeepSeek-R1-0528-Qwen3-8B-FP8-Dynamic \
--port 8000 --host 0.0.0.0

Creation

This model was created with llm-compressor by running the code snippet below.

Evaluation

TBA

Base Model

This model is a quantized version of deepseek-ai/DeepSeek-R1-0528-Qwen3-8B.

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

# Load the quantized model
model_id = "JamAndTeaStudios/DeepSeek-R1-0528-Qwen3-8B-FP8-Dynamic"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
 model_id,
 torch_dtype=torch.float16,
 device_map="auto"
)

# Example usage
messages = [
 {"role": "user", "content": "What is the capital of France?"}
]

text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(text, return_tensors="pt").to(model.device)

with torch.no_grad():
 outputs = model.generate(
 **inputs,
 max_new_tokens=100,
 temperature=0.7,
 do_sample=True
 )

response = tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)
print(response)

License

This model is released under the MIT License.