Sulat.com
AI models
NanoGPT logo

Model details

Veiled Calla 12B

Veiled Calla 12B draws from the llama family architecture to fulfill a distinct creative mission: serving as an enigmatic storytelling partner that prioritizes atmosphere over straightforward assistance. The model is engineered for immersive roleplay experiences where mystery, subtle emotional undertones, and the gradual unfolding of unspoken truths drive the narrative. Its design places strong emphasis on character consistency, ensuring personalities remain believable and grounded throughout extended interactions. This architectural lineage provides a solid foundation for crafting rich, moonlit scenarios that bloom with evocative detail, making it particularly effective in intimate and introspective settings where veiled meanings between characters come alive.

The model's training approach reveals a dual-behavior design that adapts its responses based on context. In dedicated roleplay environments like SillyTavern, it operates uncensored to support unfiltered creative exploration. Without a specific system prompt, it reverts to standard assistant-style refusals, maintaining conventional boundaries. Notably, Veiled Calla 12B has found practical application beyond standalone use—it served as a component model in the DARE TIES merge technique used to create the gama-12b model, demonstrating its utility as building blocks for more complex systems. This adaptability makes it suitable for creators seeking atmospheric narrative tools, while its proven role in model merges highlights its value in the broader ecosystem of language model development.

NanoGPTsoob3123/Veiled-Calla-12Bllama

Quick Info

Powered by
Provider
NanoGPT
Model key
soob3123/Veiled-Calla-12B
Release date
Apr 13, 2025
Last updated
Apr 13, 2025
Input modalities
Output modalities
Capabilities

Cost

Input token cost
$0.30
Output token cost
$0.30

Limits

Input tokens
32,768 tokens
Output tokens
8,192 tokens
Context window
32,768 tokens

Transparent token rates

Compare llama pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about Veiled Calla 12B

No articles yet. Fetch the latest news to show it here.

Videos about Veiled Calla 12B

More models around Veiled Calla 12B