Live leaderboards.
Each dataset below opens the same three views — the model panel, the harness panel, and the full score table — rendered live from the benchmark data behind it.
Entries marked Private are password-protected and will ask for a password before they open.
Capabilities
The six TRACES capabilities
T
Tools
Selecting, calling and correctly interpreting external tools
R
Repair
Locating and correcting its own errors once feedback arrives
A
Alternatives
Laying out competing hypotheses and keeping or discarding them as evidence accumulates
C
Coherence
Holding state, constraints and logic intact across a long chain of work
E
Evidence
Grounding every conclusion in observation, data, experiment or citation
S
Scope
Stating the conditions under which a conclusion holds, and where it does not apply