Downloads · 30 days
9
17% of all-time downloads
Arsenic-Shahrazad-12B-v4.5
Lambent/Arsenic-Shahrazad-12B-v4.5
Arsenic-Shahrazad-12B-v4.5 is a machine learning model from Lambent. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as cc-by-nc-4.0.
All-time downloads
53
Public
Parameters
12.2B
24.5 GB on disk
Likes
4
Public
Repo makeup
Click a slice to open those files.
.safetensors24.5 GB · 100%
At a glance
- License
- cc-by-nc-4.0
- Model type
- mistral
- Access
- Public
- Created
- Aug 18, 2026
- Updated
- Aug 18, 2026
- SHA
- fbed58e3
Base models
Model card
From the Hugging Face model README
Noticed she was having trouble holding multiple creative instructions in mind. Trained on GRPO with a bank of constraints generated from real prose, plus some shuffling, judge-verified. Included an auxiliary loss term based on DPO for writing quality, though length bias may have dominated and fought against length constraint. (Being aggressive enough with GRPO verification, length constraint won, hopefully.)
She seems better at handling multiple instructions now, and scores higher on creative writing overall according to Gemini 3 Flash Preview.
- Type
- mistral
- License
- cc-by-nc-4.0
- Created
- Aug 18, 2026
- Updated
- Aug 18, 2026