Instructions to use xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k with Transformers:

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

pipe = pipeline("text-generation", model="xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k")
model = AutoModelForCausalLM.from_pretrained("xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k", 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 xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k

SGLang

How to use xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k 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 "xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k" \
    --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": "xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k",
		"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 "xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k" \
        --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": "xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k with Docker Model Runner:
```
docker model run hf.co/xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k
```

xxx777xxxASD commited on Mar 5, 2024

Commit

305d132

verified ·

1 Parent(s): 82ea643

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -0

README.md CHANGED Viewed

@@ -11,6 +11,7 @@ tags:
 Test merge. Attempt to get good at RP, ERP, general things model with 128k context. Every model here has [Epiculous/Fett-uccine-Long-Noodle-7B-120k-Context](https://huggingface.co/Epiculous/Fett-uccine-Long-Noodle-7B-120k-Context) in merge instead of regular MistralYarn 128k. The reason is because i belive Epiculous merged it with Mistral Instruct v0.2 to make first 32k context expereance as perfect as possible until we reach YaRN from 32 to 128k, if not - it's sad D:, or, i get something wrong.
 [EXL2 4.0](https://huggingface.co/xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k-exl2-bpw-4.0)
 [GGUF](https://huggingface.co/xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k-GGUF)
 Here is the "family tree" of this model, im not writing full model names cause they long af

 Test merge. Attempt to get good at RP, ERP, general things model with 128k context. Every model here has [Epiculous/Fett-uccine-Long-Noodle-7B-120k-Context](https://huggingface.co/Epiculous/Fett-uccine-Long-Noodle-7B-120k-Context) in merge instead of regular MistralYarn 128k. The reason is because i belive Epiculous merged it with Mistral Instruct v0.2 to make first 32k context expereance as perfect as possible until we reach YaRN from 32 to 128k, if not - it's sad D:, or, i get something wrong.
 [EXL2 4.0](https://huggingface.co/xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k-exl2-bpw-4.0)
 [GGUF](https://huggingface.co/xxx777xxxASD/NeuralKunoichi-EroSumika-4x7B-128k-GGUF)
 Here is the "family tree" of this model, im not writing full model names cause they long af