日本語のみで絞り込む
Python implementation of TurboQuant (Zandieh et al., ICLR 2026) with original engineering findings on KV cache compression for real-world LLMs. This is a ...
This is an independent open-source implementation of the TurboQuant algorithm. All credit for the algorithm design, theoretical analysis, and original research ...
An unofficial, end-to-end PyTorch implementation of the Google Research ICLR 2026 paper: [TurboQuant: Online Vector Quantization with Near-optimal Distortion ...
2026/3/28 -I implemented Google's TurboQuant paper (KV cache compression to 3-4 bits). Repo: https://github.com/OmarHory/turboquant.
TurboQuant: KV Cache Compression for LLM Inference. Implementation of TurboQuant KV cache compression (ICLR 2026, arXiv:2504.19874) with vLLM integration.
TurboQuant compresses LLM KV caches 4-7x at inference time using random rotation + optimal scalar quantization, with near-zero quality loss. No training, no ...
2026/3/28 -Step 1: Environment Setup and W&B Initialization · Step 2: Algorithmic Implementation with PyTorch · Step 3: Hardware Diagnostics with ...
2026/3/29 -Spent ~2 days implementing this paper: TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate.
Key Ideas · Random rotation — makes any vector's coordinates statistically predictable · Scalar quantization per coordinate — enabled by near-independence after ...
3日前 -TurboQuant Pro is a broader, retrieval-first platform that uses that quantizer as one component. Documentation & reproducibility. Documentation ...