Lambent/Arsenic-Shahrazad-12B-v4.3.2

TEXT GENERATIONConcurrency Cost:1Model Size:12BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:May 20, 2026License:cc-by-nc-4.0Architecture:Transformer0.0K Open Weights Cold

Lambent/Arsenic-Shahrazad-12B-v4.3.2 is a 12 billion parameter language model with a 32768 token context length, developed by Lambent. This iteration features additional Reinforcement Learning from Human Feedback (RLHF) steps and refinements, judged by Gemma 4 31B. It demonstrates a slight increase in creative writing capabilities, particularly in generating image prompt descriptions.

Loading preview...

Overview

Lambent/Arsenic-Shahrazad-12B-v4.3.2 is a 12 billion parameter language model with a substantial 32768 token context length. This version represents an iterative improvement, incorporating additional Reinforcement Learning from Human Feedback (RLHF) steps and further refinements. The model's performance was evaluated using Gemma 4 31B as the sole judge for its RLHF process.

Key Capabilities

  • Enhanced Creative Writing: The model shows a slight improvement in creative writing abilities, specifically noted by Gemini 3 Flash.
  • Image Prompt Generation: It exhibits creativity in generating self-portrait image prompts, as interpreted by image generators.

Good For

  • Applications requiring creative text generation.
  • Tasks involving the generation of descriptive image prompts.
  • Use cases where a model with a large context window and refined RLHF training is beneficial.

Finetunes
1 models