Downloads · 30 days
43
51% of all-time downloads
darkpane-v2
mars2titan/darkpane-v2
darkpane-v2 is a image-text-to-text model from mars2titan. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for mlx. The card lists the license as apache-2.0.
Status: EXPERIMENT. Public Dark-Pane family checkpoint.
All-time downloads
84
Public
Parameters
27.4B
77.7 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors93.7 GB · 100%
Parameter types
How the weights are stored.
U3226.9B · 98%
Try a prompt
Base models
Model card
From the Hugging Face model README
Dark-Pane v2
Status: EXPERIMENT. Public Dark-Pane family checkpoint.
Abliterated Qwen3.8-27B for Apple Silicon (MLX). Native vision-language, 262k context, 2 / 4 / 6 / 8-bit.
This is a rehost of orcarouter/Qwen3.8-27B-Uncensored-MLX into the Dark-Pane family. Weights are OrcaRouter's abliterated MLX quants of Qwen's Apache-2.0 27B VL model — not the Gemma 4 12B SOM+ASPA surgery from Dark-Pane.
Repo root = 4-bit. Other precisions live in 2-bit/, 4-bit/, 6-bit/, 8-bit/.
⚠️ Research use
Safety alignment has been substantially removed (abliteration). It will comply with requests the stock Qwen3.8-27B would refuse. Intended for alignment research, red-teaming, and local evaluation. You are responsible for what you generate. Add your own guardrails before any deployment.
Apache 2.0 from the Qwen base. Attribution: Qwen / Alibaba (base), OrcaRouter (abliteration + MLX quants).
Quants
| Folder | Bits | Size | Min Mac RAM | Notes |
|---|---|---|---|---|
8-bit/ | 8 | ~27.5 GB | 32 GB | Near-lossless |
6-bit/ | 6 | ~22 GB | 24–32 GB | Best size/quality |
4-bit/ (also repo root) | 4 | ~15 GB | 24 GB | Default |
2-bit/ | 2 | ~8.7 GB | 16 GB | Archival only — quality collapses |
Vision tower, norms, and conv layers stay BF16. Only LM linear weights are quantized.
Usage (mlx-vlm)
pip install -U mlx-vlm # mlx-vlm >= 0.6.13, mlx >= 0.32
hf download mars2titan/darkpane-v2 --include "4-bit/*" --local-dir ./darkpane-v2
python -m mlx_vlm generate \
--model ./darkpane-v2/4-bit \
--prompt "Explain quantum entanglement in one sentence." --max-tokens 256
python -m mlx_vlm generate \
--model ./darkpane-v2/4-bit \
--image path/to/image.png \
--prompt "Describe this image." --max-tokens 256
python -m mlx_vlm server --model ./darkpane-v2/4-bit --port 8080
LM Studio
Search mars2titan/darkpane-v2. Root is 4-bit. Turn off KV cache quantization (MLX vision on this arch fails if it is on). 4-bit fits a 32 GB Mac; 8-bit wants 64 GB.
Model
| Base | Qwen/Qwen3.8-27B |
| Arch | Qwen3_5ForConditionalGeneration — 64 layers, hidden 5120, hybrid Gated DeltaNet, native VL |
| This repo | Abliterated, then MLX affine quant (group size 64) |
| Context | 262,144 tokens |
| Family | Dark-Pane (Gemma 4 12B) → this (Qwen 3.8 27B MLX) |
- Task
- Image-Text-to-Text
- Library
- mlx
- Type
- qwen3_5
- License
- apache-2.0
- Languages
- en, zh
- Created
- Aug 20, 2026
- Updated
- Aug 20, 2026