ForSureTesterSim/Big-Randy-NSFW-Beta-14B

TEXT GENERATIONPricing:Input $0.48 / Output $0.96Concurrent Unit Cost:1Model Size:14BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 17, 2026Architecture:Transformer Featherless Exclusive Cold

ForSureTesterSim/Big-Randy-NSFW-Beta-14B is a 14 billion parameter language model based on the Qwen3-14B-Base architecture, created by ForSureTesterSim through a DELLA merge. This model integrates components from Huihui-Qwen3-14B-abliterated-v2 and Dhanishtha-nsfw, specifically optimized for NSFW content generation. It features a 32768-token context length, making it suitable for extended, specialized text generation tasks.

Loading preview...

Model Overview

ForSureTesterSim/Big-Randy-NSFW-Beta-14B is a 14 billion parameter language model developed by ForSureTesterSim. It is built upon the Qwen/Qwen3-14B-Base architecture and was created using the DELLA merge method.

Key Capabilities

This model is a merge of two distinct models, indicating a specialized focus:

  • Content Generation: The inclusion of HelpingAI/Dhanishtha-nsfw suggests a primary optimization for generating Not Safe For Work (NSFW) content.
  • Merged Architecture: It combines huihui-ai/Huihui-Qwen3-14B-abliterated-v2 and HelpingAI/Dhanishtha-nsfw using specific density and weight parameters (0.5 and 0.4 density respectively, with a weight of 1.0 for both) to achieve its specialized output.
  • Context Length: With a context length of 32768 tokens, it can handle extensive inputs and generate longer, coherent responses.

Use Cases

Given its specific merge components, this model is primarily intended for applications requiring:

  • Specialized NSFW text generation.
  • Exploration of merged model performance in niche content domains.

Merge Details

The model was produced using mergekit with the DELLA merge method. The configuration included int8_mask: true and normalize: true parameters, with bfloat16 dtype and union tokenizer source, ensuring a robust and specialized merge.