Instructions to use jeiku/Orthocopter_8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jeiku/Orthocopter_8B with Transformers:

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

pipe = pipeline("text-generation", model="jeiku/Orthocopter_8B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("jeiku/Orthocopter_8B")
model = AutoModelForCausalLM.from_pretrained("jeiku/Orthocopter_8B")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use jeiku/Orthocopter_8B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jeiku/Orthocopter_8B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jeiku/Orthocopter_8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/jeiku/Orthocopter_8B

SGLang

How to use jeiku/Orthocopter_8B 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 "jeiku/Orthocopter_8B" \
    --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": "jeiku/Orthocopter_8B",
		"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 "jeiku/Orthocopter_8B" \
        --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": "jeiku/Orthocopter_8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use jeiku/Orthocopter_8B with Docker Model Runner:
```
docker model run hf.co/jeiku/Orthocopter_8B
```

GGFU Q8 Please.

by Ransss - opened May 10, 2024

Discussion

Ransss

May 10, 2024

I tested your models. and I like how they are special in the respond. you doing great work With your private data.

Ransss

May 10, 2024

I tried to quantize it, using GGUF-my-repo, but it didn't work.

Lewdiculous

May 10, 2024

I will try to convert this one.

Lewdiculous

May 10, 2024

@Ransss Alright, Q8 on the way before the rest since that's the one you wanted.

Should be there in a few minutes.

https://huggingface.co/Lewdiculous/Orthocopter_8B-GGUF-Imatrix

Ransss

May 10, 2024

Thank you.@Lewdiculous

jeiku

Owner May 10, 2024

•

edited May 10, 2024

I tried to quantize it, using GGUF-my-repo, but it didn't work.

Unfortunately that space does not run the bpe vocab command necessary to convert Llama 3 models. In the future, I suggest simply using llama.cpp on your local hardware or setting up a colab notebook with the appropriate scripts. Alternatively, you can wait for Lewdiculous, who I will contact for IQ quants if I deem the model worthwhile.

jeiku changed discussion status to closed May 10, 2024

jeiku changed discussion status to open May 10, 2024

jeiku

Owner May 10, 2024

@Lewdiculous thanks a bunch man, not sure how often I'll produce a new model, but may consider another one soon depending on leaderboard results.

Lewdiculous

May 10, 2024

Keep them coming.

[This discussion can be closed if you want.]

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