ToastyPigeon/gemma-4-12b-marvin-gutenberg-rp-v2

TEXT GENERATIONPricing:Input $1.2 / Cached $0.24 / Output $4.8Concurrent Unit Cost:1Model Size:12BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 15, 2026Architecture:Transformer0.0K Featherless Exclusive Cold

ToastyPigeon/gemma-4-12b-marvin-gutenberg-rp-v2 is a 12 billion parameter language model with a 32768 token context length, developed by ToastyPigeon. This model is fine-tuned for roleplay and creative writing tasks, leveraging datasets like mimo-rp, kimi-stories-instruct, and gutenberg-sft. Its primary strength lies in generating narrative content and engaging in roleplay scenarios, making it suitable for applications requiring imaginative text generation.

Loading preview...

Overview

ToastyPigeon/gemma-4-12b-marvin-gutenberg-rp-v2 is a 12 billion parameter language model designed for advanced roleplay and creative writing. It features a substantial context length of 32768 tokens, allowing for extended and coherent narrative generation. The model has been fine-tuned by ToastyPigeon on a diverse set of datasets specifically curated for enhancing its storytelling and interactive capabilities.

Key Capabilities

  • Enhanced Roleplay: Optimized for engaging in dynamic and context-rich roleplay scenarios.
  • Creative Writing: Excels at generating imaginative stories, dialogues, and narrative content.
  • Long Context Understanding: Benefits from a 32768 token context window, enabling the creation of longer, more consistent texts.

Good For

  • Interactive Storytelling Applications: Ideal for chatbots or AI companions that require deep narrative engagement.
  • Content Generation: Suitable for generating creative prose, scripts, or fictional narratives.
  • Roleplay Simulations: Can be used in virtual environments or games requiring AI characters with rich conversational abilities.

Training Datasets

The model's fine-tuning includes several specialized datasets:

Finetunes
1 models