← LiveBench leaderboards

Theory Of Mind

LiveBench · 28 models

RankModelScore
1gpt-5.4-xhigh88.462
2gpt-5.5-xhigh84.615
3claude-opus-4-6-thinking-auto-high-effort82.692
3gpt-5.5-high82.692
5claude-opus-4-7-xhigh-effort80.769
5claude-sonnet-5-xhigh-effort80.769
5deepseek-v4-pro80.769
5gemini-3.1-pro-preview-high80.769
5gemini-3.5-flash-high80.769
5gpt-5.4-nano-xhigh80.769
11claude-fable-5-xhigh-effort78.846
11claude-opus-4-5-20251101-thinking-64k-high-effort78.846
11claude-opus-4-8-xhigh-effort78.846
11gpt-5.2-2025-12-11-high78.846
11gpt-5.2-codex78.846
11gpt-5.578.846
11qwen3.7-max78.846
18gpt-5.4-mini-xhigh76.923
18minimax-m376.923
20glm-5.275
20kimi-k2.6-thinking75
22claude-sonnet-4-6-thinking-auto-medium-effort73.077
22deepseek-v4-flash73.077
24grok-build-0.169.231
24kimi-k2.7-code69.231
26qwen3.6-plus67.308
27qwen3.6-27b65.385
28grok-4.361.538