prithivMLmods/Explora-0.6B

Hugging Face
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 16, 2025License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Warm

Explora-0.6B by prithivMLmods is a 0.8 billion parameter general-purpose reasoning model, fine-tuned on Qwen3-0.6B with a 32768 token context length. It is specifically optimized for science and code-focused reasoning tasks, leveraging the Open-Omega-Explora-2.5M dataset. This model excels at symbolic reasoning, basic computations, and structured logic, making it ideal for exploratory workflows in STEM domains and low-resource environments.

Loading preview...

Explora-0.6B: A Compact Reasoning Model for STEM

Explora-0.6B, developed by prithivMLmods, is a lightweight and efficient 0.8 billion parameter language model built upon the Qwen3-0.6B architecture. It has been fine-tuned using the first 100,000 entries of the Open-Omega-Explora-2.5M dataset, which comprises a diverse collection of math, code, and science problems. This specialized training makes Explora-0.6B particularly adept at general-purpose reasoning within STEM fields.

Key Capabilities

  • General-Purpose STEM Reasoning: Handles symbolic reasoning, basic computations, and structured logic with clarity, specifically for code and science problems.
  • Balanced Thinking Mode: Supports moderate reasoning depth, aiming to provide step-by-step problem-solving, function generation, and explanatory outputs while minimizing hallucination.
  • Output Flexibility: Capable of generating responses in various formats including Markdown, Python, JSON, or plain text, suitable for both human readability and integration into automated pipelines.
  • Compact & Deployable: With 0.8 billion parameters, it is designed for efficient deployment in offline environments, low-resource inference setups, and educational tools requiring fast, reliable logic.

Good For

  • Educational and lightweight research tools.
  • General science and programming assistance.
  • Low-resource STEM assistants for code labs or classrooms.
  • Fast-response agents for structured reasoning tasks, especially with specific and scoped questions.