Downloads · 30 days
18
6% of all-time downloads
zynxware/nexus-prime
nexus-prime is a text generation model from zynxware. Use it when you need the model to write or continue text. The card lists the license as mit.
PROPERLY CONFIGURED MODEL - CLEAN OUTPUT
Downloads · 30 days
18
6% of all-time downloads
All-time downloads
314
Public
Parameters
81.9M
825 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors328 MB · 100%
From the Hugging Face model README
PROPERLY CONFIGURED MODEL - CLEAN OUTPUT
from transformers import pipeline
# Simple usage - NO GARBAGE OUTPUT
generator = pipeline('text-generation',
model='zynxware/nexus-prime',
max_new_tokens=80,
temperature=0.7,
repetition_penalty=1.2)
prompt = "User: Hello\nNEXUS-PRIME:"
response = generator(prompt)[0]['generated_text']
# Extract clean response
if "NEXUS-PRIME:" in response:
clean_response = response.split("NEXUS-PRIME:")[-1].strip()
print(clean_response)