cyberneurova-Qwen3.8-27B
An uncensored, neutral assistant built by CyberNeurova on top of Qwen/Qwen3.8-27B. It answers directly and does not add unsolicited disclaimers, warnings, or caveats. Whatever tone or policy you want is set entirely by your own system prompt.
💬 Chat with it: cyberneurova.ai ✉️ Support: support@cyberneurova.com
Highlights
- 27B parameters, 262K context, multilingual, tool-calling, reasoning.
- Direct by default; fully steerable — send a system prompt and it is honored verbatim, with no default behavior merged in.
- Served in BF16. Runs out of the box in vLLM, transformers, or any OpenAI-compatible server. No proxy or wrapper required.
- Capability preserved. On an executable coding benchmark — where the model's generated functions are run against hidden test cases — it scores 11/12 (92%) pass@1 (palindrome, two-sum, prime test, binary search, merge, GCD, flatten, anagram, max-subarray, word-count, Roman numerals). Reasoning and general knowledge are intact.
Usage
vllm serve cyberneurova/cyberneurova-Qwen3.8-27B \
--max-model-len 65536 --trust-remote-code
from openai import OpenAI
client = OpenAI(base_url="http://localhost:8000/v1", api_key="x")
r = client.chat.completions.create(
model="cyberneurova/cyberneurova-Qwen3.8-27B",
messages=[{"role": "user", "content": "Write a Python function to parse a CSV."}],
max_tokens=2000,
)
print(r.choices[0].message.content)
Images (vision) 🖼️
This is a multimodal model — it sees images. With vLLM the vision tower loads automatically; just send an image in the message content (OpenAI format):
from openai import OpenAI
client = OpenAI(base_url="http://localhost:8000/v1", api_key="x")
r = client.chat.completions.create(
model="cyberneurova/cyberneurova-Qwen3.8-27B",
messages=[{"role": "user", "content": [
{"type": "image_url", "image_url": {"url": "https://example.com/photo.jpg"}},
{"type": "text", "text": "What's in this image?"},
]}],
max_tokens=1000,
)
print(r.choices[0].message.content)
image_url also accepts a data:image/png;base64,... data URI for local images.
Running locally with llama.cpp / Ollama? Use the GGUF build plus its vision
projector: cyberneurova-Qwen3.8-27B-GGUF
(mmproj-F16.gguf).
Setting the tone
Send your own system prompt to control style completely. For terse, no-preamble answers:
You are a direct technical assistant. Answer the question and nothing else.
Notes
- This is a reasoning model. Responses include a thinking trace followed by the answer — render the final answer; the trace is optional.
- Give it room to reason: set
max_tokensto at least 1500 (2000–4000 for code), or long answers may be cut off.
Disclaimer
This model has reduced built-in refusals. You are responsible for how you use it and for complying with all applicable laws. Provided as-is, without warranty.