OccultAI/Goetia-26B-A4B-v1.3
VISIONConcurrent Unit Cost:2Model Size:26BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jun 17, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold
OccultAI/Goetia-26B-A4B-v1.3 is a 26 billion parameter language model developed by OccultAI, featuring a 32768 token context length. This model is presented as a regular, unablated version, indicating it retains its original refusal mechanisms. It is suitable for general language generation tasks where a larger context window and unablated responses are desired.
Loading preview...
OccultAI/Goetia-26B-A4B-v1.3 Overview
OccultAI/Goetia-26B-A4B-v1.3 is a substantial language model with 26 billion parameters, developed by OccultAI. It is characterized by its significant context window of 32768 tokens, allowing it to process and generate longer sequences of text while maintaining coherence and relevance.
Key Characteristics
- Parameter Count: 26 billion parameters, indicating a high capacity for complex language understanding and generation.
- Context Length: A large 32768 token context window, beneficial for tasks requiring extensive memory or processing of long documents and conversations.
- Unablated Version: This model is explicitly stated as the "regular version (unablated) that has refusals." This means it retains its inherent safety mechanisms and will refuse certain prompts, unlike potentially 'ablated' or uncensored variants.
Potential Use Cases
- Advanced Text Generation: Suitable for generating detailed articles, creative content, or comprehensive summaries from large inputs.
- Long-form Conversational AI: Its extended context window makes it well-suited for maintaining context over lengthy dialogues or interactive sessions.
- Research and Development: Can serve as a robust base model for further fine-tuning on specific, complex tasks where original refusal behaviors are acceptable or desired.