Instructions to use DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567 with Transformers:

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

pipe = pipeline("text-generation", model="DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567")
model = AutoModelForMultimodalLM.from_pretrained("DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567")
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 DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567

SGLang

How to use DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567 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 "DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567" \
    --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": "DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567",
		"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 "DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567" \
        --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": "DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567 with Docker Model Runner:
```
docker model run hf.co/DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Mixtral-6x7B-Instruct-v0.1 (bfloat16)

The Mixtral-6x7B-Instruct-v0.1 model is a derivative of the mistralai/Mixtral-8x7B-Instruct-v0.1 model. It was created by selectively trimming the original model and retaining only the 0th, 2nd, 4th, 5th, 6th, and 7th experts from each layer.

The trimming process was facilitated by the Mixtral-Expert-Trimmer tool, developed specifically for this purpose.

The model is still in testing phase. It is not clear whether it works.

License

The Mixtral-6x7B-Instruct-v0.1 model is open-source and licensed under the Apache 2.0 License. For more information, please refer to the LICENSE file.

Feeling Generous? 😊

Eager to buy me a cup of 2$ coffe or iced tea?🍵☕ Sure, here is the link: https://ko-fi.com/drnicefellow. Please add a note on which one you want me to drink?

Downloads last month: 88

Safetensors

Model size

35B params

Tensor type

BF16

Model tree for DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567

Quantizations

2 models

Collection including DrNicefellow/Mixtral-6x7B-Instruct-v0.1-bfloat16-Trimmed024567

Trimmed-Mixtral-instruct

Collection

1 item • Updated Apr 6, 2025