duizhang
duizhang is an open-source CLI framework for evaluating large language model outputs in the context of Chinese couplet generation. It provides tools for NLP researchers to assess and analyze LLM performance, including RAG support and couplet-specific metrics.
Open SourceGPL-3.0-or-laterCLI
1star
4features
2026since
Overview
4 featuresduizhang sits in PulseGate's LLM eval & observability category. It focuses on evaluating the quality of LLM-generated Chinese couplets for NLP research. duizhang is an open-source project aimed at NLP researchers and developers working with LLMs and Chinese text. duizhang is open source under the GPL-3.0-or-later license. It ships for the command line.
cycleuser builds and maintains duizhang, and it first shipped in 2026. Development happens publicly on GitHub with 4 commits in the last 90 days. Key capabilities include LLM evaluation, chinese couplet analysis, and RAG support.
Summary written by a language model from the project’s public pages.
- LLM evaluation
- Chinese couplet analysis
- RAG support
- CLI interface
Tags
llm-evaluationchinese-coupletrag-framework
AI capabilities
Text
Built with & integrations
Framework
flask
AI providers
local_oss
Runs on
CLI
Trust & compliance
License
GPL-3.0-or-later
Verified signals
Indexing history
1What PulseGate has recorded for this listing
Frequently asked questions about duizhang
- What does duizhang do?
- Duizhang focuses on evaluating the quality of LLM-generated Chinese couplets for NLP research. It is catalogued under LLM eval & observability on PulseGate.
- Who should use duizhang?
- duizhang is an open-source project built for NLP researchers and developers working with LLMs and Chinese text.
- Does duizhang have a free plan?
- Yes — duizhang is open source under the GPL-3.0-or-later license and free to use.
- What platforms does duizhang run on?
- duizhang runs on the command line.
- Is duizhang still active?
- The GitHub repository shows 4 commits in the last 90 days.
- What are alternatives to duizhang?
- Similar projects tracked by PulseGate include constraintloop, trajectory-judge, and DeFake.constraintlooptrajectory-judgeDeFake
- Who makes duizhang?
- duizhang is developed by cycleuser.
- When did duizhang launch?
- duizhang first shipped in 2026.
Also in LLM eval & observability
Same category — not a similarity match