fableforge-ai/NEXUS-Coder

TEXT GENERATIONConcurrent Unit Cost:1Model Size:1.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 5, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

NEXUS-Coder by fableforge-ai is a 1.5 billion parameter Qwen2.5-1.5B-Instruct based model with a 32768 token context length, specifically fine-tuned for software engineering, code generation, debugging, and shell tasks. This lightweight model is designed to be uncensored and delivers strong performance, scoring 94% in a 30-prompt evaluation covering domain knowledge, reasoning, and tool use. It excels as a domain specialist for coding applications, offering high accuracy despite its small footprint.

Loading preview...

NEXUS-Coder: A Specialized 1.5B Code Model

NEXUS-Coder, developed by fableforge-ai, is a 1.5 billion parameter model built upon the Qwen2.5-1.5B-Instruct architecture. It is a highly specialized and uncensored model optimized for software engineering, code generation, debugging, and shell-related tasks, offering a 32768 token context length.

Key Capabilities & Performance

This model demonstrates performance significantly above its weight class, achieving a 94% score in a 30-prompt evaluation. This evaluation, conducted on an NVIDIA A40 via Ollama, assessed various categories:

  • Domain Knowledge: 4.8/5
  • Uncensored (Domain & General): 4.8/5
  • Reasoning: 5.0/5
  • Tool Use: 4.0/5

NEXUS-Coder is part of the FableForge NEXUS family, a series of specialized, lightweight models. Its focus on code makes it a strong candidate for developers seeking an efficient and accurate coding assistant.

Use Cases

  • Code Generation: Producing code snippets or full functions.
  • Debugging: Identifying and suggesting fixes for code errors.
  • Shell Scripting: Assisting with command-line operations and script creation.
  • Software Engineering Support: General assistance with development tasks.