gohumanize/gohumanize-open-humanizer

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Sep 18, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

The GoHumanize Open Humanizer is a 4 billion parameter model, a full fine-tune of Qwen3-4B by GoHumanize, designed to transform AI-generated English prose into more natural, human-like writing. It was trained on 2,957 pairs of AI-styled and human-original passages, including texts from Project Gutenberg and US federal agencies. This model excels at restructuring sentences and varying prose style to remove common AI patterns, making it ideal for humanizing machine-generated content.

Loading preview...

GoHumanize Open Humanizer Overview

The GoHumanize Open Humanizer is a 4 billion parameter model, a full fine-tune of Qwen3-4B, developed by GoHumanize. Its primary function is to rewrite AI-styled English text into more natural, human-like prose. This model was trained on a unique dataset of 2,957 pairs of AI-generated and human-written passages, sourced from public domain books (Project Gutenberg) and modern US federal agency documents. It serves as a public research and educational tool to demonstrate the principles behind AI text humanization systems.

Key Capabilities

  • Text Humanization: Rewrites AI-generated text to improve sentence length variation, use concrete wording, establish natural rhythm, and eliminate filler transitions.
  • Meaning Preservation: Designed to maintain the original meaning, facts, and order of ideas during the rewriting process.
  • Restructuring: Learns to restructure sentences rather than just swapping words, aiming for a more profound stylistic change.
  • Fact Retention: While generally good, users should verify numbers and facts as some rewrites may rephrase or occasionally lose numerical data.

Good for

  • Educational Purposes: Demonstrating the methodology for developing AI text humanization systems.
  • Stylistic Rewriting: Transforming machine-generated content to sound more natural and less robotic.
  • Short Passages: Optimized for passages between 80 and 300 words; longer documents should be processed paragraph by paragraph.

Limitations

  • The model's style leans towards literary (pre-1929 books) and official (US government writing), with casual modern writing being a gap.
  • Not intended to reproduce the architecture or performance of GoHumanize.ai's production systems.
  • Makes no claims about bypassing AI detectors.
  • English only.

Finetunes
3 models