The 12 approaches I tested before finding one...

I keep seeing ML papers that only show the final method. No dead ends, no "we tried X and it was a disaster." Just polished results on a polished pipeline. This is the opposite of that. Here is a complete record of every approach I tested building NexusQuant, a KV cache compressor for LLMs. Including the ones that failed spectacularly. The problem I was trying to solve KV cache is what limits LLM context windows. A 70B model on a single A100 can handle roughly 128K tokens before running out of m