
LLMLingua
Microsoft's prompt-compression method: drops low-information tokens; the repo reports up to 20x compression.
Website captured Oct 19

Microsoft's prompt-compression method: drops low-information tokens; the repo reports up to 20x compression.
Website captured Oct 19
Fast BPE tokenizer for OpenAI models: count tokens before they hit the context window.
Latest collected value
Latest collected value
Latest collected value
Latest collected value
Latest collected value
Latest collected value
Latest collected value
| Signal | LLMLingua 7d | LLMLingua 28d | tiktoken 7d | tiktoken 28d |
|---|---|---|---|---|
| GitHub stars | ▲0.2% | ▲0.9% | ▲0.2% | ▲0.9% |
| GitHub forks | ▲0.7% | ▲2.5% | ▲0.7% | ▲1.7% |
| PyPI downloads (last week) | ▲21.1% | ▲15.0% | ▲5.5% | ▲12.7% |
| Hacker News stories (7 days) | — | — | — | — |
| Hacker News points (7 days) | — | — | — | — |
| GitHub open issues & PRs | ▲0.0% | ▼0.8% | ▲0.7% | ▲4.2% |
| GitHub contributors | ▲0.0% | ▲0.0% | ▲0.0% | ▲4.0% |