Instructions to use blascotobasco/Qwen3-Next-256E-Abliterated-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use blascotobasco/Qwen3-Next-256E-Abliterated-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="blascotobasco/Qwen3-Next-256E-Abliterated-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("blascotobasco/Qwen3-Next-256E-Abliterated-Instruct")
model = AutoModelForCausalLM.from_pretrained("blascotobasco/Qwen3-Next-256E-Abliterated-Instruct")
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

vLLM

How to use blascotobasco/Qwen3-Next-256E-Abliterated-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "blascotobasco/Qwen3-Next-256E-Abliterated-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "blascotobasco/Qwen3-Next-256E-Abliterated-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/blascotobasco/Qwen3-Next-256E-Abliterated-Instruct

SGLang

How to use blascotobasco/Qwen3-Next-256E-Abliterated-Instruct 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 "blascotobasco/Qwen3-Next-256E-Abliterated-Instruct" \
    --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": "blascotobasco/Qwen3-Next-256E-Abliterated-Instruct",
		"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 "blascotobasco/Qwen3-Next-256E-Abliterated-Instruct" \
        --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": "blascotobasco/Qwen3-Next-256E-Abliterated-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use blascotobasco/Qwen3-Next-256E-Abliterated-Instruct with Docker Model Runner:
```
docker model run hf.co/blascotobasco/Qwen3-Next-256E-Abliterated-Instruct
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

WARNING: This model is broken and should not be used.

Experimental prune of Qwen3 Next 80B A3B from 512 experts to 256 experts, using HuiHui's abliterated version as a base.

In my testing the model performs well, but I welcome feedback.

Update after more testing: This model is a little bit stupid. Use with caution.

Downloads last month: 7

Safetensors

Model size

42B params

Tensor type

BF16

Model tree for blascotobasco/Qwen3-Next-256E-Abliterated-Instruct

Base model

Qwen/Qwen3-Next-80B-A3B-Instruct

Finetuned

(34)

this model

Finetunes

1 model

Quantizations

3 models