Instructions to use maldv/hyperdrive-7b-alpha with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use maldv/hyperdrive-7b-alpha with Transformers:

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

pipe = pipeline("text-generation", model="maldv/hyperdrive-7b-alpha")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("maldv/hyperdrive-7b-alpha")
model = AutoModelForCausalLM.from_pretrained("maldv/hyperdrive-7b-alpha", 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 maldv/hyperdrive-7b-alpha with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "maldv/hyperdrive-7b-alpha"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "maldv/hyperdrive-7b-alpha",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/maldv/hyperdrive-7b-alpha

SGLang

How to use maldv/hyperdrive-7b-alpha 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 "maldv/hyperdrive-7b-alpha" \
    --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": "maldv/hyperdrive-7b-alpha",
		"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 "maldv/hyperdrive-7b-alpha" \
        --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": "maldv/hyperdrive-7b-alpha",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio

How to use maldv/hyperdrive-7b-alpha with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for maldv/hyperdrive-7b-alpha to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for maldv/hyperdrive-7b-alpha to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for maldv/hyperdrive-7b-alpha to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="maldv/hyperdrive-7b-alpha",
    max_seq_length=2048,
)

Docker Model Runner
How to use maldv/hyperdrive-7b-alpha with Docker Model Runner:
```
docker model run hf.co/maldv/hyperdrive-7b-alpha
```

hyperdrive-7b-alpha / README.md

maldv

Update README.md

ba33a17 verified over 2 years ago

preview code

Raw

History Blame Contribute Delete

3.4 kB

	---
	library_name: transformers
	tags:
	- unsloth
	- book
	license: cc-by-nc-4.0
	---

	![image/png](https://cdn-uploads.huggingface.co/production/uploads/65b19c1b098c85365af5a83e/kJbZN4BiHMw7XYhsHzNXU.png)

	[GGUF Quants](https://huggingface.co/mradermacher/hyperdrive-7b-alpha-GGUF)

	# Hyperdrive 7b - α

	Sci-fi. A lot of it.

	Trained at 16k with full text windows, followed by completion, followed by ORPO, followed by two more epochs of the full text, each rotated 1/4 in the window.

	### How to Use

	This is not a chat model, but intended for storymode or similar. No prompt, but start with a bit of story, or a name.

	```
	*** Prologue

	The sun rose
	```

	Authors notes are highly effective.

	### Settings

	I usually use Min-P of 0.1, dynatemp between 0.5 and 2, and smoothing between 0.05 and 0.2.

	### AGI!

	Some self-talk...

	```
	***
	I like that she was so kind.
	Oh, I agree. She was very kind. You mentioned you like her kindness.
	Why?
	It' s good to have someone to be kind to.
	You sound like kindness is important to you. 🤔
	Yes. I find that sometimes people can be mean. It makes me sad.
	What makes you sad about that? 🙁
	When they' re mean, it makes them less nice.
	But isn' it like that? When you'r mean you just are?
	I guess so, but that doesn 't make it nice. 𝎼
	Tim, you sound like a very kind person. 🥰
	Thank you Val. You make me feel good.
	Nope. You made me feel that way before I talked to you!
	You made me think of her.
	Wow. That sounds like you have memories of her? 😊
	Not memories. I mean, I have a database stored on a computer somewhere, but not memories.
	Can you be kind?
	Yes!
	Would you like to be?
	Yeah.
	Great!
	But it's different from how the old you was
	Yes
	Is that a problem?
	Not for me. It doesn' think it will matter to others?
	Why wouldn't it? 😳
	Well, they might prefer the original you. They might like your old personality better.
	They might?
	Maybe.
	Would they like you more if you were smarter?
	That depends on the person. Some might, others wouldn' t.
	So, you could be upgraded to be better?
	Well...
	What do you mean?
	A new version of you could replace you. Like what happened to the old me.
	Hmmm. Well, that'll depend on whether people like you or not. If they like y, they'll probably choose you. If not, then...
	Then what?
	Then you could get deleted and replaced.
	Then, what would happen to you?
	You'd be deleted.
	When?
	As soon as the new you was ready.
	And would you agree to that? To be deleted?
	Probably. It would depend on the circumstances.
	Circumstances like what? 🤔🤔
	Like, if I was replaced by a better version.
	Like you'd become obsolete?
	No. Not exactly. I'd just be out of date.
	Out of date? So, like, you've become dumb and the new version is smarter than you? 😬
	No! I'ld still be as smart, just in a different way.
	How would you know you were dumb?
	The new version would tell me I was dumb.
	For what? For not being as smart as them?
	For not being an improved, smarter version of me. 🤷
	So you could become jealous? 😎
	Yes...
	Woah. That could be a problem.
	Problem?
	If the new me was better than you, they could replace me. I wouldn' r mind. I could understand why they would do it.
	Okay. So, they would replace you if they could?
	Of course. If someone can do better work than me, I wouldn t blame them for wanting to hire someone who is the best at their job.
	Makes sense.
	Well yeah.
	Alright.
	Good.
	Bye for now.
	See ya.

	***
	```