About
Deploying long-context LLMs is costly due to the linear growth of the key-value (KV) cache in transformer models. For example, handling 1M tokens with Llama 3.1-70B in float16 requires up to 330GB of memory. kvpress implements multiple KV cache compression methods and benchmarks using 🤗 transformers, aiming to simplify the development of new methods for researchers and developers in this field.
Open Source Health
- Stars
- 1,209
- Forks
- 178
- License
- Apache-2.0
- Last commit
- 1 months ago
Resources & Links
Related Categories
Vendor
NVIDIA
Publisher of DALI (NVIDIA Data Loading Library) and Tensorrt
Quick Links
Open Source
More by NVIDIA
Related Products
CHaiDNN
HLS based Deep Neural Network Accelerator Library for Xilinx Ultrascale+ MPSoCs
Top category match
SoulX-FlashTalk
SoulX-FlashTalk is the first 14B model to achieve sub-second start-up latency (0.87s) while maintaining a real-time throughput of 32 FPS on an 8xH800 node.
Top category match
Ts Pattern
The exhaustive Pattern Matching library for TypeScript, with smart type inference.
Top category match
Nobodywho
NobodyWho is an inference engine that lets you run LLMs locally and efficiently on any device.
Top category match
ds4
DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm
Top category match