Instructions to use v2ray/DeepSeek-V3-FP16-Atten-NaN with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use v2ray/DeepSeek-V3-FP16-Atten-NaN with Transformers:

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

pipe = pipeline("text-generation", model="v2ray/DeepSeek-V3-FP16-Atten-NaN")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("v2ray/DeepSeek-V3-FP16-Atten-NaN", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use v2ray/DeepSeek-V3-FP16-Atten-NaN with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "v2ray/DeepSeek-V3-FP16-Atten-NaN"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "v2ray/DeepSeek-V3-FP16-Atten-NaN",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/v2ray/DeepSeek-V3-FP16-Atten-NaN

SGLang

How to use v2ray/DeepSeek-V3-FP16-Atten-NaN 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 "v2ray/DeepSeek-V3-FP16-Atten-NaN" \
    --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": "v2ray/DeepSeek-V3-FP16-Atten-NaN",
		"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 "v2ray/DeepSeek-V3-FP16-Atten-NaN" \
        --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": "v2ray/DeepSeek-V3-FP16-Atten-NaN",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use v2ray/DeepSeek-V3-FP16-Atten-NaN with Docker Model Runner:
```
docker model run hf.co/v2ray/DeepSeek-V3-FP16-Atten-NaN
```

DeepSeek V3 FP16 Atten NaN

This is a minimal reproduceable sample to let the final layer of DeepSeek V3's attention output NaNs when using data type float16.

Run the run.py to see the NaNs.

Weights are converted to bfloat16 using the original float8 e4m3fn, then converted to float16, then extracted from the final layer's attention.

Downloads last month: -; Downloads are not tracked for this model. How to track

Model tree for v2ray/DeepSeek-V3-FP16-Atten-NaN

Base model

deepseek-ai/DeepSeek-V3

Finetuned

(87)

this model