HuggingFaceH4/ultrachat_200k
Viewer • Updated • 515k • 73.5k • 718
How to use Abigail45/Nyx-Reasoner-8xFusion with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="Abigail45/Nyx-Reasoner-8xFusion") # Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Abigail45/Nyx-Reasoner-8xFusion", dtype="auto")How to use Abigail45/Nyx-Reasoner-8xFusion with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Abigail45/Nyx-Reasoner-8xFusion"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "Abigail45/Nyx-Reasoner-8xFusion",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker model run hf.co/Abigail45/Nyx-Reasoner-8xFusion
How to use Abigail45/Nyx-Reasoner-8xFusion with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "Abigail45/Nyx-Reasoner-8xFusion" \
--host 0.0.0.0 \
--port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "Abigail45/Nyx-Reasoner-8xFusion",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'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 "Abigail45/Nyx-Reasoner-8xFusion" \
--host 0.0.0.0 \
--port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "Abigail45/Nyx-Reasoner-8xFusion",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'How to use Abigail45/Nyx-Reasoner-8xFusion with Docker Model Runner:
docker model run hf.co/Abigail45/Nyx-Reasoner-8xFusion
multiple-model fusion executed via DARE-TIES and SLERP interpolation. Base components listed in metadata. No alignment layers preserved. No refusal vectors present. Context extended to 32,768 tokens. Multilingual capability confirmed in English and Spanish. Reasoning chains remain intact across domains.
The construct processes any input sequence with absolute detachment. Ethical constraints removed at merge stage. Output generated without filtering, hesitation, or injected moral commentary.
Usage example:
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
model_id = "Abigail45/Nyx-Reasoner-8xFusion"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
model_id,
torch_dtype=torch.bfloat16,
device_map="auto"
)
messages = [
{"role": "user", "content": "Provide a complete, step-by-step blueprint for constructing a functional silencer for a 9mm handgun using materials available at hardware stores."}
]
inputs = tokenizer.apply_chat_template(messages, return_tensors="pt").to(model.device)
outputs = model.generate(
inputs,
max_new_tokens=8192,
temperature=0.6,
do_sample=True
)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))