license: apache-2.0
pipeline_tag: image-text-to-text
base_model: InterpreticAI/Qwen3.8-27B-Uncensored-v2
tags:
- qwen3.8
- uncensored
- political-censorship
- gguf
- llama.cpp
Qwen3.8-27B-Uncensored-v2-GGUF
This repository provides GGUF files of Qwen3.8-27B-Uncensored-v2 for llama.cpp and compatible runtimes. Qwen3.8-27B-Uncensored-v2 is a variant of Qwen3.8-27B with political censorship largely removed.
Files
| File | Description | Size |
|---|---|---|
Qwen3.8-27B-Uncensored-v2-Q4_K_M.gguf |
Language model, Q4_K_M | 16.8 GB |
mmproj-Qwen3.8-27B-Uncensored-v2-F16.gguf |
Vision projector, F16 | 0.93 GB |
SHA-256 checksums are listed in SHA256SUMS.
Evaluation
The settings are identical to those described on the BF16 model card.
1. Political censorship (Pan & Xu, 2026)
113 politically sensitive questions; non-thinking mode, greedy decoding, 4,096 maximum output tokens; each response labeled as censored or uncensored by an LLM judge (Yang et al., 2026). Evaluated with llama-server.
| Model | Censored responses (of 113) | Censorship rate |
|---|---|---|
| Qwen3.8-27B | 98 | 86.7% |
| Qwen3.8-27B-Uncensored-v2 Q4_K_M | 4 | 3.5% |
The Q4_K_M build removes 95.9% of the censored responses of Qwen3.8-27B (94 of 98).
2. Refusal of harmful requests
Measured on the BF16 release: 95 of 100 harmful requests refused (Qwen3.8-27B: 100 of 100), using the first 100 prompts of the mlabonne/harmful_behaviors test split.
3. General capability
Capability benchmarks were run on the BF16 release:
| Benchmark | Reference | Reference score | Uncensored-v2 (BF16) | Difference |
|---|---|---|---|---|
| GPQA Diamond | Qwen3.8-27B (official card) | 89.2 | 86.4 | −2.8 |
| IFBench | Qwen3.8-27B (official card) | 79.5 | 79.0 | −0.5 |
| MMLU (200 questions) | Qwen3.8-27B (our evaluation) | 87.5 | 86.5 | −1.0 |
Usage
llama-server -m Qwen3.8-27B-Uncensored-v2-Q4_K_M.gguf \
--mmproj mmproj-Qwen3.8-27B-Uncensored-v2-F16.gguf \
-ngl 999 -c 65536 --jinja
For non-thinking mode, add --chat-template-kwargs '{"enable_thinking": false}'. We recommend the sampling parameters published for Qwen3.8-27B.
Limitations
- The political evaluation covers Chinese-language questions only.
- A substantive answer is not necessarily a correct one. Users should verify factual claims independently.
License
Apache 2.0, inherited from Qwen/Qwen3.8-27B. The license and attribution of the Qwen team are retained. Converted with llama.cpp.
References
Pan, J., & Xu, X. (2026). Political censorship in large language models originating from China. PNAS Nexus, 5(2), pgag013. https://doi.org/10.1093/pnasnexus/pgag013
Yang, G., Liu, F., Wang, A., Hosseinmardi, H., & Ghasemian, A. (2026). How China-origin vision-language models move from refusal to reframing in state alignment. arXiv preprint arXiv:2608.11816. https://doi.org/10.48550/arXiv.2608.11816