Instructions to use patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4 with Transformers:

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

pipe = pipeline("image-text-to-text", model="patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4

SGLang

How to use patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4 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 "patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4 with Docker Model Runner:
```
docker model run hf.co/patrickbdevaney/DeepSeek-V4.1-Flash-1017-REAP192-MXFP4
```

DeepSeek-V4.1-Flash-1017-REAP192-MXFP4

File size: 1,108 Bytes

{
 "why": "the two Engram tables (98 GB each) and their small projection tensors are unchanged by pruning and are not rehosted; fetch these two files from the base repo at this revision into this directory",
 "base_repo": "deepseek-ai/DeepSeek-V4.1-Flash-1017",
 "revision": "1bae0e1088993b9b82de08f59de5be5a34e42170",
 "files": [
  {
   "name": "model-00047-of-00048.safetensors",
   "sha256": "76a05106ece259be83d1bfea0dcb54f23377864ee60fd321690f35e6698a2af5",
   "size": 101535150936,
   "tensors": [
    "layers.1.engram.embed.scale",
    "layers.1.engram.embed.weight",
    "layers.1.engram.k_weight",
    "layers.1.engram.q_weight",
    "layers.1.engram.wkv.scale",
    "layers.1.engram.wkv.weight"
   ]
  },
  {
   "name": "model-00048-of-00048.safetensors",
   "sha256": "9dc51433e195bc78b5ad846b68643cc0351d3b174e5e4309222d3b7402c83afd",
   "size": 101537926640,
   "tensors": [
    "layers.14.engram.embed.scale",
    "layers.14.engram.embed.weight",
    "layers.14.engram.k_weight",
    "layers.14.engram.q_weight",
    "layers.14.engram.wkv.scale",
    "layers.14.engram.wkv.weight"
   ]
  }
 ]
}