mamii76/Snowball-8B-Abliterated1

TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:8kTool Calling:SupportedPublished:Jun 3, 2026Architecture:Transformer Cold

mamii76/Snowball-8B-Abliterated1 is an 8 billion parameter language model created by mamii76, utilizing a Passthrough merge method based on deepseek-ai/DeepSeek-R1-Distill-Llama-8B. This model integrates the full layer range of its base model, offering a foundational language model with an 8192 token context length. It is suitable for general language generation tasks where a merged architecture derived from DeepSeek's R1-Distill-Llama is beneficial.

Loading preview...

Overview

mamii76/Snowball-8B-Abliterated1 is an 8 billion parameter language model developed by mamii76. It was created using the mergekit tool, specifically employing a "Passthrough" merge method. This approach means the model is directly based on a single foundational model, rather than combining multiple distinct models.

Key Characteristics

  • Base Model: The model's architecture and weights are derived from deepseek-ai/DeepSeek-R1-Distill-Llama-8B.
  • Merge Method: Utilizes the Passthrough merge method, effectively adopting the base model's structure directly.
  • Layer Integration: The merge configuration specifies the inclusion of all 32 layers from the deepseek-ai/DeepSeek-R1-Distill-Llama-8B model.
  • Context Length: Features an 8192 token context window.

Use Cases

This model is suitable for applications requiring a robust 8B parameter language model with the characteristics inherited from the DeepSeek-R1-Distill-Llama architecture. It can be used for various natural language processing tasks, including text generation, summarization, and question answering, leveraging its foundational capabilities.