This model is a general-purpose fine-tune designed for high flexibility across several domains. It is intended for research, creative applications, and tasks requiring complex output.
-
Uncensored Nature: This model is described as uncensored. Users should be aware that the model may generate content that is sensitive, explicit, biased, or generally harmful, as safety filters have been removed or significantly relaxed.
-
Responsible Use: Due to its uncensored nature, this model is explicitly not recommended for public-facing deployments without aggressive post-processing and external safety layers.
1models:
2 - model: janhq/Jan-v1-2509
3 parameters:
4 density: 0.2
5 weight: 0.2
6 - model: huihui-ai/Huihui-Qwen3-4B-Thinking-2507-abliterated
7 parameters:
8 density: 0.5
9 weight: 0.5
10 - model: minchyeom/Qwaifu
11 parameters:
12 density: 0.3
13 weight: 0.3
14merge_method: ties
15base_model: janhq/Jan-v1-2509
16parameters:
17 normalize: true
18dtype: float16
19
1!pip install -qU transformers accelerate
2
3from transformers import AutoTokenizer
4import transformers
5import torch
6
7model = "bunnycore/Qwen3-4B-Max-Ties"
8messages = [{"role": "user", "content": "What is a large language model?"}]
9
10tokenizer = AutoTokenizer.from_pretrained(model)
11prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
12pipeline = transformers.pipeline(
13 "text-generation",
14 model=model,
15 torch_dtype=torch.float16,
16 device_map="auto",
17)
18
19outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
20print(outputs[0]["generated_text"])
