Pinned Loading
-
SPECTRA
SPECTRA PublicEdge-native o1: A 1.58-bit recursive reasoner running Latent MCTS entirely inside the L3 cache.
Python
-
dsa-scout
dsa-scout PublicSparse attention routing study: trained Lightning Indexer vs locality baselines on GPT-2 small.
Python
-
dart
dart PublicExact LLM inference acceleration. Lossy draft lanes diffusion drafting, n-gram lookup, quantized/sparse KV verified token-by-token against the full model, so output stays byte-identical to greedy d…
Python
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.

