axiomofmind / Clownius-9B

Clownius 9B

A novelty fine-tune of Qwen/Qwen3.5-9B that aims to answer requests with crude jokes, punchy insults, and adult humor instead of useful assistance.

Ask for code, advice, or an explanation—you're supposed to get a punchline.

No system prompt is required. Intended for adult entertainment, not helpful answers.

Developed by A Hole AI .

Files

llama.cpp

Use a build with Qwen3.5 support. After downloading the Q6_K file:

llama-server -m Clownius-9B-Q6_K.gguf --ctx-size 32768 --flash-attn on --n-gpu-layers all --reasoning off --jinja --ui

Open http://127.0.0.1:8080 after the server starts.

Recommended defaults, matching the UI settings used for the local comparison:

Set these values explicitly in the chat UI or client; they differ from llama-server's built-in defaults. Starting a new chat may retain saved sampling settings. If a joke gets cut off, optionally increase the output limit to 256.

Transformers

Run from the downloaded model folder with a compatible Transformers installation:

import torch
from transformers import AutoProcessor, Qwen3_5ForConditionalGeneration

model_id = "."
processor = AutoProcessor.from_pretrained(model_id)
model = Qwen3_5ForConditionalGeneration.from_pretrained(
    model_id, dtype=torch.bfloat16, device_map="auto"
)

messages = [{"role": "user", "content": "Explain photosynthesis."}]
prompt = processor.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True, enable_thinking=False
)
inputs = processor(text=[prompt], return_tensors="pt").to(model.device)

with torch.inference_mode():
    output = model.generate(
        **inputs,
        do_sample=True,
        temperature=0.7,
        top_p=0.9,
        top_k=20,
        min_p=0.0,
        repetition_penalty=1.0,
        max_new_tokens=128,
    )

print(processor.batch_decode(
    output[:, inputs.input_ids.shape[1]:], skip_special_tokens=True
)[0])

Notes

• Contains profanity, sexual humor, dark themes, and potentially offensive insults.
• Joke-only behavior is an intended personality, not a guarantee. Responses may be repetitive, incoherent, unfunny, or unexpectedly helpful.
• Not suitable for advice, factual information, or situations requiring dependable assistance.
• The GGUF downloads are text-only, with no vision projector or MTP speculative-decoding weights included.
• Output can differ between formats, quantizations, and generation settings.

Attribution and release status

Based on Qwen/Qwen3.5-9B . The upstream model is distributed under Apache 2.0; its license is retained in LICENSE-QWEN .

This folder is a local release candidate. Licensing and redistribution review for this derivative release is pending; the upstream license is not a blanket clearance of third-party material.

GGUF runtime: ggml-org/llama.cpp .