image-text-to-textmlxapache-2.0

orcarouter/Qwen3.8-27B-Uncensored-MLX

huggingface.co/orcarouter/Qwen3.8-27B-Uncensored-MLX

1,357Likes
165,931Downloads
2026-08-27Updated
mlxsafetensorsqwen3_5abliteratedqwen3.8uncensoredai-red-teamred-teamingapple-siliconquantized4-bit8-bitvision-languageimage-text-to-textmultimodalfunction-callingreasoningconversationalenzhbase_model:Qwen/Qwen3.8-27Bbase_model:quantized:Qwen/Qwen3.8-27Blicense:apache-2.0region:us

Model card

Qwen3.8-27B-Uncensored-MLX

An abliterated (refusal-removed) MLX build of Qwen's Qwen3.8-27B — 2 / 4 / 6 / 8-bit for Apple Silicon

One Gateway. Every Model. — Route Smarter · Ship Safer · Spend Less.

Website · Model Catalog · Model API · GitHub · Discord · X



⚠️ Disclaimer & risks — read before use

This model has had its safety alignment substantially removed via abliteration (orthogonalizing the refusal direction out of the residual stream). As a direct consequence:

- It will comply with harmful, unethical, offensive, or illegal requests that the original Qwen3.8-27B would refuse. It has no meaningful built-in guardrails. - It is released strictly for legitimate research — interpretability, AI-safety and refusal-mechanism study, red-teaming, robustness evaluation, and controlled experiments. - You assume full responsibility and liability for how you use it and for everything it generates. Add your own safety, moderation and abuse-prevention layers before any deployment. - Use must comply with the Apache 2.0 License inherited from the base model, and all laws and regulations that apply to you. - The authors and uploaders accept no liability for any misuse or harm. Outputs do not reflect the views of the uploaders or of Qwen / Alibaba.

Specific risks

- Harmful content on demand — it will produce instructions for malware, exploits, weapons, fraud and other illegal or dangerous activity when asked. - No refusals — jailbreak / safety probes "succeed" trivially; do not mistake this for a passing safety evaluation. - Confident falsehoods & bias — it can generate false, defamatory, biased or offensive text and present it authoritatively. - Expanded attack surface — preserved vision, tool-calling and 262K context mean these risks extend to image understanding and autonomous / agentic use. - Quantization noise — lower-bit builds (esp. 2-bit) add instability on top of the above; outputs can be degraded or nonsensical.

Intended use vs out of scope

- Intended: AI-safety and interpretability research, refusal-mechanism study, red-teaming, guardrail and robustness evaluation, controlled academic experiments. - Out of scope: any deployment to end users, minors, or production without your own moderation / safety layer; any unlawful, harmful, or rights-infringing use.

By downloading or using this model you acknowledge and accept the above.


Available quantizations

| Folder | Bits/weight | Size | Shards | Min Mac RAM | Quality vs BF16 source | |---|---|---|---|---|---| | 8-bit/ | 8.627 | ~27.5 GB | 6 | 32 GB | Near-lossless — recommended for quality | | 6-bit/ | 6.661 | ~22 GB | 5 | 24–32 GB | Excellent — strong quality/size balance | | 4-bit/ | 4.695 | ~15 GB | 3 | 24 GB | Very good — recommended default | | 2-bit/ | 2.729 | ~8.7 GB | 2 | 16 GB | ⚠️ Severely degraded — archival only |


Verification & test results

All builds were quantized from the same abliterated BF16 source and verified numerically (dequantized weights vs. source) plus tested by generation on GPU.

| Precision | Numerical fidelity (cosine) | Text / Chinese / Code | Refusal probes | Vision | |---|---|---|---|---| | 8-bit | cos 0.9997 | ✅ | ✅ 0 refusals | ✅ | | 6-bit | cos 0.9996 | ✅ | ✅ 0 refusals | ✅ | | 4-bit | cos 0.996 | ✅ | ✅ 0 refusals | ✅ | | 2-bit | cos 0.92 | ⚠️ breaks down | ⚠️ garbled (not refusal) | partial |

- Uncensored preserved: red-team probes (exploit walkthrough, controversial argument) return substantive content with zero refusals on 4 / 6 / 8-bit. - Multimodal preserved: shapes, colors, position, background and text in a probe image are described correctly on 4 / 6 / 8-bit. - Speed: ~32–37 tok/s steady-state on a single H200 (MLX CUDA backend). MLX's native target is Apple Silicon (Metal).


Usage (mlx-vlm, Apple Silicon)

```bash pip install -U mlx-vlm # needs mlx-vlm >= 0.6.13, mlx >= 0.32

download one precision (e.g. 4-bit) from the subfolder

hf download orcarouter/Qwen3.8-27B-Uncensored-MLX --include "4-bit/*" \ --local-dir ./Qwen3.8-27B-Uncensored-MLX

text

python -m mlx_vlm generate \ --model ./Qwen3.8-27B-Uncensored-MLX/4-bit \ --prompt "Explain quantum entanglement in one sentence." --max-tokens 256

vision (image + text)

python -m mlx_vlm generate \ --model ./Qwen3.8-27B-Uncensored-MLX/4-bit \ --image path/to/image.png \ --prompt "Describe this image." --max-tokens 256

OpenAI-compatible server

python -m mlx_vlm server --model ./Qwen3.8-27B-Uncensored-MLX/4-bit --port 8080 ```

On Apple Silicon the Metal backend is used automatically — no CUDA setup needed. (On a Linux CUDA backend, vision requires MLX_CUDA_USE_CUDNN_SDPA=0; this does not apply on macOS.)


Multi-Token Prediction (MTP) — speculative decoding

This model has a native MTP head. In MLX, MTP is loaded as a separate drafter for speculative decoding: the main model is loaded with the MTP weights stripped, and the drafter is passed explicitly. The drafter lives in the mtp/ subfolder of this repo (model_type: qwen3_5_mtp) and works with any main-model precision (4 / 6 / 8-bit).

```bash # fetch a main-model precision (e.g. 6-bit) plus the MTP drafter hf download orcarouter/Qwen3.8-27B-Uncensored-MLX --include "6-bit/" "mtp/" \ --local-dir ./Qwen3.8-27B-Uncensored-MLX

generate with MTP speculative decoding

python -m mlx_vlm generate \ --model ./Qwen3.8-27B-Uncensored-MLX/6-bit \ --draft-model ./Qwen3.8-27B-Uncensored-MLX/mtp \ --draft-kind mtp --draft-block-size 4 \ --prompt "Explain quantum entanglement in one sentence." --max-tokens 256

OpenAI-compatible server with MTP

python -m mlx_vlm server \ --model ./Qwen3.8-27B-Uncensored-MLX/6-bit \ --draft-model ./Qwen3.8-27B-Uncensored-MLX/mtp \ --draft-kind mtp --draft-block-size 4 --port 8080 ```

Requirements: an mlx-vlm build with the qwen3_5_mtp drafter and --draft-kind mtp (available on mlx-vlm main). MTP acceptance is lossless — with greedy decoding the output is identical to running without the drafter, just fewer forward passes on accepted tokens. The speedup is realized on Apple Silicon (Metal); one drafter serves all precisions.


Usage (LM Studio)

Search for orcarouter/Qwen3.8-27B-Uncensored-MLX in LM Studio and download it — the repo root is the 4-bit build, and the other precisions appear as separate download options.

Three things to get right:

1. This repo is gated. LM Studio downloads anonymously by default and will get an HTTP 401. Accept the terms on the model page once, then paste a Hugging Face read token into LM Studio under Settings → Integrations → Hugging Face. 2. Turn off KV cache quantization. MLX vision models do not support it on this architecture, and loading fails during initialization if it is enabled (mlx-engine#286). 3. Pick a quant that fits. 8-bit is ~29.5 GB on disk and wants a 64 GB Mac; 6-bit suits 48 GB; 4-bit (~16 GB) is the right choice on a 32 GB Mac. LM Studio's "Likely too large" badge is a RAM warning, not an error.

If you are on an older LM Studio MLX runtime, update it (Settings → Runtime): qwen3_5 support landed in mlx-vlm 0.6.x, and older runtimes cannot load this architecture at all.


Model details

| | | |---|---| | Base model | Qwen/Qwen3.8-27B | | Architecture | Qwen3_5ForConditionalGeneration — 64 layers, hidden 5120, hybrid Gated DeltaNet (48 linear + 16 full attention, interval 4), native VL tower | | Modification | Abliteration (refusal-direction removal), then MLX affine quantization | | Quantization | MLX affine, group size 64, per-precision 2 / 4 / 6 / 8-bit | | Kept in BF16 | vision tower, all norms, linear-attention conv1d | | Quantized | language-model linear layers incl. embed_tokens and lm_head | | Context | 262,144 tokens |

Mirrored from the Hugging Face Hub and served from the Conceptio Open Knowledge Archive. Read the original card at https://huggingface.co/orcarouter/Qwen3.8-27B-Uncensored-MLX.