Near-optimal KV cache compression for LLM inference — TurboQuant (vector quantization) + TriAttention (spectral compression)
[email protected] is safe to use (health: 61/100)
Get this data programmatically — free, no authentication.
curl https://depscope.dev/api/check/pypi/aither-kvcacheLast updated · 2026-04-15T00:35:53.752193Z