3 ms·
Sqz: Compress LLM context to save tokens and reduce costs
- sea-gold 6mo agosqz vs rtk — Token Savings Benchmark: https://github.com/ojuschugh1/sqz/blob/main/docs/benchmark-vs-rtk.md https://github.com/ojuschugh1/sqz/blob/main/docs/benchmark-v...