Instructions to use MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8 with Transformers:

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

pipe = pipeline("text-generation", model="MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8")
model = AutoModelForCausalLM.from_pretrained("MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8", device_map="auto")
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 Settings

vLLM

How to use MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8

SGLang

How to use MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8 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 "MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8" \
    --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": "MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8",
		"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 "MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8" \
        --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": "MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8 with Docker Model Runner:
```
docker model run hf.co/MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8
```

Quantized using the default exllamav3 (0.0.3) quantization process.

Original model: https://huggingface.co/nbeerbower/Yanfei-v2-Qwen3-32B
exllamav3: https://github.com/turboderp-org/exllamav3

Yanfei-v2-Qwen3-32B

A repair of Yanfei-Qwen-32B by TIES merging huihui-ai/Qwen3-32B-abliterated, Zhiming-Qwen3-32B, and Menghua-Qwen3-32B using mergekit.

Sponsorship

This model was made possible with compute support from Nectar AI. Thank you! ❤️

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: ./Zhiming-Qwen3-32B-merged
    parameters:
      weight: 1
      density: 1
  - model: ./Menghua-Qwen3-32B-merged
    parameters:
      weight: 1
      density: 1
  - model: huihui-ai/Qwen3-32B-abliterated
    parameters:
      weight: 1
      density: 1
merge_method: ties
base_model: nbeerbower/Yanfei-Qwen3-32B
parameters:
  weight: 1
  density: 1
  normalize: true
  int8_mask: true
dtype: bfloat16

Downloads last month: 7

Safetensors

Model size

17B params

Tensor type

F16

I16

Model tree for MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8

Base model

nbeerbower/Yanfei-v2-Qwen3-32B

Quantized

(7)

this model

Dataset used to train MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8

Paper for MetaphoricalCode/Yanfei-v2-Qwen3-32B-exl3-8bpw-hb8

Resolving Interference When Merging Models

Paper • 2306.01708 • Published Jun 2, 2023 • 19