Fork-and-upgrade experiments on top of NX-AI TFLA mLSTM Triton kernels.
Automatic xLSTM architecture configuration, driven by task type and live GPU memory profiling.
LinearTile: Transformers → Ψ(N) — O(N) linear attention with mathematical Psi guarantee
A retention-based language model built from scratch in C: O(N) training, O(1) inference
YAN (Yet Another Notation) parser for Python
Anthropic, PBC
Visionary sponsor
Bloomberg
Visionary sponsor
Hudson River Trading
Visionary sponsor
Meta
Visionary sponsor
NVIDIA
Visionary sponsor
Microsoft
Sustainability sponsor
Depot
Continuous Integration
AWS
Cloud computing and Security Sponsor
Datadog
Monitoring
Fastly
CDN
Google
Download Analytics
Sentry
Error logging
StatusPage
Status page