Search Results for "key value"

vLLM is a fast and easy-to-use library for LLM inference and serving. High-throughput serving with various decoding algorithms, including parallel sampling, beam search, and more.

R-KV is an open-source research project that focuses on improving the efficiency of large language model inference through key-value cache compression techniques. Modern transformer models rely heavily on KV caches during autoregressive decoding, which store intermediate attention states to accelerate generation. However, these caches can consume large amounts of memory, especially in reasoning-oriented models with long context windows. R-KV introduces a method for compressing the KV cache during decoding, allowing models to maintain reasoning performance while reducing memory consumption and computational overhead. ...

KVCache-Factory is an open-source research framework designed to explore and implement unified key-value cache compression techniques for autoregressive transformer models. In large language models, the key-value cache stores intermediate attention states that enable efficient token generation during inference, but these caches can consume large amounts of GPU memory when handling long contexts. KVCache-Factory provides a platform for implementing and evaluating multiple compression strategies that reduce memory usage while preserving model performance. ...

...Traditional retrieval-augmented generation systems rely on real-time retrieval of documents from databases or vector stores during inference. CAG proposes a different approach by preloading relevant knowledge into the model’s context window and precomputing the model’s key-value cache before queries are processed. This strategy allows the model to generate responses using the cached context directly, eliminating the need for repeated retrieval operations during runtime. As a result, the approach can significantly reduce latency and simplify system architecture compared with traditional RAG pipelines. ...

LMCache is an extension layer for LLM serving engines that accelerates inference, especially with long contexts, by storing and reusing key-value (KV) attention caches across requests. Instead of rebuilding KV states for repeated or shared text segments, LMCache persists and retrieves them from multiple tiers—GPU memory, CPU DRAM, and local disk—then injects them into subsequent requests to reduce TTFT and increase throughput. Its design supports reuse beyond strict prefix matching and enables sharing across serving instances, improving efficiency under real multi-tenant traffic. ...

...The project is built around the idea that a model can be trained in a parallelizable way like a GPT-style transformer while running inference with recurrent efficiency. This gives RWKV important advantages for long-context use, including lower memory pressure and no traditional key-value cache requirement. The repository includes training code, model notes, research material, and references to current RWKV weights. Its main value is providing the foundation for experimenting with efficient large language models that combine transformer-like scalability with RNN-like runtime behavior.

...It supports a wide range of augmentation techniques, allows to easily combine these and to execute them in random order or on multiple CPU cores, has a simple yet powerful stochastic interface and can not only augment images but also key points/landmarks, bounding boxes, heatmaps and segmentation maps. Affine transformations, perspective transformations, contrast changes, gaussian noise, dropout of regions, hue/saturation changes, cropping/padding, blurring, etc. Rotate image and segmentation map on it by the same value sampled. Convert keypoints to distance maps, extract pixels within bounding boxes from images, clip polygon to the image plane, etc. ...