https://sachinsmc.me/blog/tags/vector-search/ Recent content in Vector-Search on Sachin Chavan Hugo en-us Tue, 23 Jun 2026 12:00:00 +0400 https://sachinsmc.me/blog/posts/turboquant-data-free-quantization/ Tue, 23 Jun 2026 12:00:00 +0400 https://sachinsmc.me/blog/posts/turboquant-data-free-quantization/ Google’s TurboQuant compresses LLM KV caches and embedding vectors to 3-4 bits with no training and no codebook, and still beats methods that need both. The clever part is not the quantizer, it is what it refuses to store. Here is the intuition, with runnable NumPy you can paste and check.