by jserv · Claude Skill · ★ 45
cjk-token-reducer Reduce Claude Code token usage by 35-50% when using CJK languages. The Problem CJK (Chinese, Japanese, Korean) languages consume 2-4x more tokens than English for the same semantic content. This discrepancy leads to higher costs, faster context exhaustion, and reduced context windows for RAG/agent workflows. Token ratios based on BPE tokenizer analysis. Actual savings depend on text complexity and technical term density.
| Stars | 45 |
| Forks | 7 |
| Language | Rust |
| Category | Claude Skill |
| License | MIT |
| Quality Score | 58.3/100 |
| Last Updated | 2026-05-07 |
| Created | 2026-01-17 |
| Platforms | claude-code, rust |
| Est. Tokens | ~40k |
These tools work well together with cjk-token-reducer for enhanced workflows:
Explore other popular claude skill tools:
cjk-token-reducer is Reduce Claude Code token usage by 35-50% when using CJK (Chinese, Japanese, and Korean). It is categorized as a Claude Skill with 45 GitHub stars.
cjk-token-reducer is primarily written in Rust. It covers topics such as cjk-tokenizer, claude-code, llm-inference.
You can find installation instructions and usage details in the cjk-token-reducer GitHub repository at github.com/jserv/cjk-token-reducer. The project has 45 stars and 7 forks, indicating an active community.
cjk-token-reducer is released under the MIT license, making it free to use and modify according to the license terms.