thunlp/InfLLM
The code of our paper "InfLLM: Unveiling the Intrinsic Capacity of LLMs for Understanding Extremely Long Sequences with Training-Free Memory"
This project helps large language models (LLMs) understand and process extremely long text inputs, like those found in complex conversations or lengthy documents. It takes an existing LLM and allows it to maintain context over vast amounts of information, producing more coherent and relevant responses than standard methods. Anyone who works with LLMs and needs them to process very long conversations, extensive reports, or continuous data streams will find this useful.
395 stars. No commits in the last 6 months.
Use this if you need an existing large language model to comprehend and respond intelligently to input sequences that are hundreds of thousands or even a million tokens long without retraining the model.
Not ideal if your primary use case involves short, concise text inputs, as the benefits of long-sequence processing won't be realized.
Stars
395
Forks
39
Language
Python
License
MIT
Category
Last pushed
Apr 20, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/thunlp/InfLLM"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
jncraton/languagemodels
Explore large language models in 512MB of RAM
microsoft/unilm
Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities
haizelabs/verdict
Inference-time scaling for LLMs-as-a-judge.
albertan017/LLM4Decompile
Reverse Engineering: Decompiling Binary Code with Large Language Models
bytedance/Sa2VA
Official Repo For Pixel-LLM Codebase