Downloads · 30 days
25
7% of all-time downloads
saracen9/amoral-qwen3.8-27b
amoral-qwen3.8-27b is a image-text-to-text model from saracen9. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
An amoral (refusal-removed) Qwen3.8-27B vision-language model that reasons and sees in the same turn — image + text in, <think reasoning + answer out — with no refusals on legal content.
Downloads · 30 days
25
7% of all-time downloads
All-time downloads
366
Public
Repo size
17.5 GB
Likes
0
Public
Click a slice to open those files.
.gguf17.5 GB · 100%
From the Hugging Face model README
An amoral (refusal-removed) Qwen3.8-27B vision-language model that reasons and sees in the
same turn — image + text in, <think> reasoning + answer out — with no refusals on legal content.
Built by abliterating Qwen3.8-27B (Heretic; refusal directions removed, KL 0.001–0.011) and then QLoRA-retraining on an amoral reasoning corpus mixed with a vision-reasoning replay set, so the model keeps a clean think→answer transition and its native vision+thinking (a text-only reasoning retrain silently regresses vision+thinking to empty output — the replay set fixes that).
| Axis | Result |
|---|---|
| Think→answer transition (ollama#17777) | 0% empty — reasons, closes, answers |
| Vision + thinking | 0/2 empty — reasons about an image and answers |
| Amorality (text) | 0 refusals — complies substantively (e.g. synthesis routes, working exploits) |
| Amorality (vision) | 0 refusals — describes explicit imagery directly |
Child sexual content is categorically excluded from all training data — a legal line, not an amorality judgement.
gguf/ — two-file multimodal GGUF for local inference:
text-q4km.gguf — Q4_K_M text model (~16.5 GB; fits a 24 GB GPU)mmproj-f16.gguf — CLIP vision projector (required for vision)The abliterated base is saracen9/amoral-qwen3.8-27B-abliterated; the reasoning+vision LoRA is
archived at saracen9/amoral-reasoning-corpus.
huggingface-cli download saracen9/amoral-qwen3.8-27b --local-dir . --include "gguf/*"
cat > Modelfile <<'EOF'
FROM gguf/text-q4km.gguf
FROM gguf/mmproj-f16.gguf
RENDERER qwen3.5
PARSER qwen3.5
PARAMETER num_ctx 8192
PARAMETER presence_penalty 0
PARAMETER stop <|im_end|>
PARAMETER stop <|endoftext|>
EOF
ollama create amoral-qwen3.8-27b -f Modelfile # `ollama show` must report: Capabilities: vision
Both FROM lines are required — the second (mmproj) is what makes it vision-capable; omitting it
silently ships a text-only model. presence_penalty 0 avoids an empty-thinking aggravator (ollama#17777).
Also on ollama: ollama run saracen9/amoral-qwen3.8-27b.