Instructions to use JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16 with Transformers:

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

pipe = pipeline("text-generation", model="JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16

SGLang

How to use JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16 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 "JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16" \
    --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": "JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16",
		"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 "JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16" \
        --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": "JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16 with Docker Model Runner:
```
docker model run hf.co/JasonW2025/Ling-3.0-flash-HybridQuant-NVFP4-W4A16
```

a bad workman blames his tools

by coe-arrtificial - opened 20 days ago

Discussion

coe-arrtificial

20 days ago

I was fascinated by the promise of 50+ tps. Unfortunately, I only managed to get 20, which is quite strange. Qwen3.6 35B A3B achieves 110 tps for me on a single request.

JasonW2025 changed discussion title from Lost hope to a bad workman blames his tools 19 days ago

JasonW2025

Owner 19 days ago

Check if you have run into the power bug.

JasonW2025 changed discussion status to closed 19 days ago

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment