fableforge-ai/NEXUS-Science

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

fableforge-ai/NEXUS-Science is a 1.5 billion parameter instruction-tuned causal language model, based on Qwen/Qwen2.5-1.5B-Instruct, specifically optimized as a domain specialist for scientific reasoning and research. This lightweight model is designed for high performance in scientific contexts, achieving a 93% score in a 30-prompt evaluation covering domain knowledge, reasoning, and tool use. It is part of the FableForge NEXUS family, offering uncensored responses within its scientific domain.

Loading preview...

NEXUS-Science: Scientific Reasoning Specialist

NEXUS-Science is a 1.5 billion parameter language model developed by FableForge, built upon the Qwen/Qwen2.5-1.5B-Instruct architecture. It is specifically fine-tuned to excel in scientific reasoning and research tasks, making it a domain specialist within the FableForge NEXUS ecosystem.

Key Capabilities & Performance

  • Domain Expertise: Achieves a high score of 4.7/5 in domain knowledge, demonstrating deep understanding in scientific fields.
  • Reasoning & Tool Use: Scores 5.0/5 in reasoning and 4.2/5 in tool use, indicating strong analytical and practical application abilities.
  • Uncensored Responses: Provides uncensored outputs for both domain-specific and general queries, scoring 5.0/5 and 4.4/5 respectively.
  • Lightweight & Efficient: Despite its small size (1.5B parameters), it "punches far above its weight," achieving a total benchmark score of 93% (140/150) in a comprehensive 30-prompt evaluation.

Part of the NEXUS Family

NEXUS-Science is one of six specialized models in the NEXUS family, each focusing on a distinct domain. This model is tailored for "Science & Research," complementing others like NEXUS-Coder (Software Engineering) and NEXUS-Medical (Medicine & Healthcare).

Good For

  • Applications requiring specialized scientific knowledge and reasoning.
  • Research assistance and data analysis in scientific fields.
  • Scenarios where a lightweight, high-performing model for scientific inquiry is needed.