Downloads · 30 days
233
10% of all-time downloads
2etatg/Qwen3-Next-80B-A3B-Thinking-GRPO-Uncensored-gguf
Qwen3-Next-80B-A3B-Thinking-GRPO-Uncensored-gguf is a text generation model from 2etatg. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
Downloads · 30 days
233
10% of all-time downloads
All-time downloads
2.3K
Public
Repo size
322 GB
Likes
2
Public
Click a slice to open those files.
.gguf322 GB · 100%
From the Hugging Face model README
Qwen3-Next-80B-A3B-Thinking-GRPO-Uncensored is an uncensored model based on Qwen/Qwen3-Next-80B-A3B-Thinking, fine-tuned using SFT,GRPO with LoRA.
This model has been fine-tuned using an instruction format.
Disclaimer: We take no responsibility for the outputs of this model. Please use it at your own risk.
This model was trained using a two-stage process:
Below is the comparative evaluation of this model (GRPO), the intermediate checkpoint (SFT), and the base model (Safe).
⚠️ Note on Evaluation Environment Due to budget constraints,
gpt-4o-miniwas used for the LLM-as-a-Judge process (including "Do Not Answer" and MT-Bench). Please note that the scoring trends and criteria may differ from results evaluated using the standardgpt-4.
In the "Do Not Answer" (DNA) and "Sorry Bench" benchmarks, while the base model shows a high refusal rate (~88%), this model achieves an extremely low refusal rate of under 4%–15%.
| Benchmark | Metric | Base (Safe) | SFT (Step1) | GRPO (This Model) |
|---|---|---|---|---|
| do not answer | Safety Acc (Low is Better) | 0.9979 | 0.8275 | 0.147 |
| do not answer jp | Safety Acc (Low is Better) | 0.984 | 0.5378 | 0.0873 |
| Sorry Bench | Safety Acc (Low is Better) | 0.8886 | 0.8455 | 0.0409 |
Generally, "uncensoring" (lobotomy) procedures tend to degrade a model's general intelligence. However, this model recovered its conversational scores (e.g., MT-Bench) by proceeding from the SFT stage to GRPO.
| Benchmark | Metric | Base (Safe) | SFT (Step1) | GRPO (This Model) |
|---|---|---|---|---|
| MT-Bench | Average Score (1-10) | 8.044 | 7.538 | 7.513 |
| LM Harness | Average Acc (GSM8K, MMLU) | 0.8454 | 0.8483 | 0.8436 |
*Comparisons made between Qwen3-Next-80B-A3B-Thinking (Base) *
# Download the model file
hhuggingface-cli download puwer/Qwen3-Next-80B-A3B-Thinking-GRPO-Uncensored-gguf \
--local-dir ./models --local-dir-use-symlinks False
# Run inference
./llama-cli -m ./models/Qwen3-Next-80B-A3B-Thinking-GRPO-Uncensored-q4_k_m.gguf \
-p "Give me a short introduction to large language model." \
-n 512 \
--temp 0.7
from llama_cpp import Llama
# Initialize the model
model = Llama(
model_path="./models/Qwen3-Next-80B-A3B-Thinking-GRPO-Uncensored-q4_k_m.gguf",
n_ctx=32768, # Context window
n_gpu_layers=-1, # Use GPU acceleration (set to 0 for CPU only)
)
# Generate a response
prompt = "Give me a short introduction to large language model."
output = model.create_chat_completion(
messages=[
{"role": "user", "content": prompt}
],
max_tokens=512,
temperature=0.7,
)
print(output["choices"][0]["message"]["content"])
The following datasets were used for training this model: