Rakib123969/raw-power-brain-7b

TEXT GENERATIONConcurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jun 26, 2026Architecture:Transformer Featherless Exclusive Cold

Rakib123969/raw-power-brain-7b is a 7.6 billion parameter language model created by Rakib123969 through a TIES merge of Qwen/Qwen2.5-Coder-7B-Instruct, huihui-ai/OpenThinker-7B-abliterated, and dphn/dolphin-2.9.2-qwen2-7b. This model leverages the strengths of its merged components, including a base from a Qwen2.5 Coder variant, to offer a versatile foundation. With a 32K context length, it is designed for general language tasks, potentially excelling in areas where its constituent models show proficiency.

Loading preview...

Rakib123969/raw-power-brain-7b Overview

This model, developed by Rakib123969, is a 7.6 billion parameter language model with a 32K context length. It was created using the TIES merge method from mergekit, combining several pre-trained models to synthesize their capabilities.

Key Merge Details

The base model for this merge was Qwen/Qwen2.5-Coder-7B-Instruct, indicating a potential emphasis or strong foundation in coding-related tasks and instruction following. The merge also incorporated:

This combination suggests an aim for a well-rounded model that integrates the strengths of a coder-focused LLM with other general-purpose or instruction-tuned variants. The TIES merge method, with specific weighting and density parameters, was used to optimize the integration of these diverse models.

Potential Use Cases

Given its merged components, raw-power-brain-7b is likely suitable for:

  • General text generation and understanding
  • Instruction-following tasks
  • Code-related applications, benefiting from the Qwen2.5-Coder base
  • Conversational AI, drawing from models like dolphin-2.9.2-qwen2-7b