sequelbox/gemma-4-12B-it-Esper4FableComposer

TEXT GENERATIONPricing:Input $1.2 / Cached $0.24 / Output $4.8Concurrent Unit Cost:1Model Size:12BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 3, 2026Architecture:Transformer0.0K Featherless Exclusive Cold

The sequelbox/gemma-4-12B-it-Esper4FableComposer is a 12 billion parameter language model based on the Google Gemma-4 architecture, created by sequelbox through a DARE TIES merge. This model integrates capabilities from 'agentic-fable' and 'Esper' models, suggesting an optimization for complex narrative generation and agentic reasoning tasks. It is designed for applications requiring advanced text composition and intelligent interaction within a 32768 token context window.

Loading preview...

Model Overview

The sequelbox/gemma-4-12B-it-Esper4FableComposer is a 12 billion parameter language model built upon the google/gemma-4-12B-it base. It was developed by sequelbox using the DARE TIES merge method, combining two specialized models: yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2 and ValiantLabs/gemma-4-12B-it-Esper4. This merging strategy aims to integrate their respective strengths into a single, more versatile model.

Key Capabilities

  • Advanced Composition: The inclusion of 'fable' and 'composer' in its lineage suggests enhanced capabilities in generating complex narratives, creative writing, and structured text.
  • Agentic Reasoning: The 'agentic' component indicates potential for improved performance in tasks requiring planning, decision-making, and multi-step problem-solving.
  • Gemma-4 Architecture: Benefits from the underlying Google Gemma-4 architecture, providing a robust foundation for general language understanding and generation.
  • Extended Context Window: Supports a context length of 32768 tokens, enabling the processing and generation of longer, more coherent texts.

Good For

  • Creative Writing & Storytelling: Ideal for generating intricate plots, character dialogues, and diverse narrative styles.
  • AI Agent Development: Suitable for building intelligent agents that require sophisticated reasoning and interaction capabilities.
  • Complex Text Generation: Applications demanding high-quality, contextually rich, and structured output over extended passages.

Finetunes
1 models