Benchmark Almanac

HUD AI

We track 2 benchmarks that HUD AI made or helped make.

0 in use 0 with public code 0 with a leaderboard here

Benchmarks

Name Kind Released Best score Status
Autonomy-10 Agents complete browser, desktop and tool-use tasks across nine domains that measure skill, flexibility and intelligence. Web browsing Jan 2025 Not in cards
Yet Another Kernel Kernel Models optimize GPU kernels for speed while a verifier rejects numerical errors and input-insensitive shortcuts. GPU kernels Nov 2026 Not in cards