Huiqiang Jiang

Hi @defatoraj, by default, using LLMLingua requires NVIDIA CUDA to be enabled. You can switch to CPU mode with the following settings. ```python from llmlingua import PromptCompressor llm_lingua = PromptCompressor(device_map="cpu")...

Hello @yenif, Thank you for your help and support. We agree that agent scenarios like AutoGen are well-suited for approaches like LLMLingua to reduce token redundancy. However, there might be...

Hi @ignorejjj, thank you for your support for our work. No problem, I will post the parameters we used in the experiment below for reference. We will also release the...

> Very appreciate your awesome work. Could you please provide the code for evaluation, including batched inference? Hi @zhyunlong, thank you for your support with LLMLingua. We utilize the same...

Hi @whm233, thank you for your support and interest in LLMLingua. Although I'm not an expert in **LangChain**, based on my experience, I believe its usage in **LangChain** should be...

Thanks to @thehapyone's contribution, LLMLingua is now available in Langchain. You can follow this notebook for guidance.

Hi @lucasjinreal, Thank you for your question. I believe that LLMLingua & LongLLMLingua can retain most of the information even when compressing very long prompts. For example, let's do a...

Hi @Avkashhirpara, you can disregard this warning. In LLMLingua, we process the data segment by segment and compress the KV cache that exceeds the context window limits.

Hi @seanzhang-zhichen, thank you for your attention and feedback on our project. This issue is caused by a bug in the tokenizer's decoding of Unicode. We will fix it soon...

> I also encountered the same problem, I want to know when this bug can be repair. Hi @Yongxin-Hu, I apologize, but due to bandwidth limits, we're currently unable to...