Compare top models

LLM Benchmarks

See how today's leading models write Spacetime code, scored task by task on live modules.

Trends

Task pass rate over time, per model.
Claude Opus 4.8Claude Sonnet 4.6DeepSeek V4 FlashDeepSeek V4 ProGemini 3.1 Pro

Task Pass Rate Over Time

Click a data point to view that day's leaderboard.