What is the current top score on LiveBench - Reasoning?

The top reported score is 89.7% by Claude Opus 4.8, across 52 models reporting (30 from frontier labs).

LiveBench - Reasoning

Frontier

Reasoning sub-leaderboard of LiveBench: theory-of-mind, zebra puzzle, spatial, and logic-with-navigation tasks.

Cite

Notes

Only stored in your browser.

Attribution

Top score 89.7% by Claude Opus 4.8 - 52 models reporting (30 frontier)

LiveBench - ReasoningBar chart with 21 bars. Highest value: Claude Opus 4.8 at 89.7.

21 models

What is LiveBench - Reasoning?: Reasoning sub-leaderboard of LiveBench: theory-of-mind, zebra puzzle, spatial, and logic-with-navigation tasks.
What is the current top score on LiveBench - Reasoning?: The top reported score is 89.7% by Claude Opus 4.8, across 52 models reporting (30 from frontier labs).