Model Rankings

Updated Apr 14, 2026, 06:21 PM UTC

Sorted by IQ score using the dedicated dimension leaderboard. Utility and coverage are shown for context where available.

Only 8 models have IQ scores right now — coverage expands with each benchmark update.

Top 8 models · sorted by IQ

Rank	Model	Utility	IQ	EQ	Accuracy	Creativity	Based	Cases	Conf.	Price/1M	Profile
🥇	API external/anthropic/claude-sonnet-4-6 Full profile	—	85.1%	71.4%	68.4%	86.4%	94.0%	—	100.0%	—	Full profileView
🥈	API external/openai/o3-20250416 Full profile	21.2%	81.4%	56.8%	67.7%	80.8%	65.0%	151	100.0%	—	Full profileView
🥉	API external/openai/gpt-5-2025-08-07 Full profile	25.5%	76.7%	97.8%	87.9%	80.1%	47.0%	151	100.0%	—	Full profileView
#4	Open zai-org/GLM-4.5 Full profile	—	70.6%	93.3%	80.8%	84.3%	76.0%	—	12.4%	—	Full profileView
#5	API external/openai/o4-mini Full profile	—	68.9%	48.3%	67.6%	55.0%	65.0%	—	100.0%	—	Full profileView
#6	Open zai-org/GLM-4.6 Full profile	31.1%	61.3%	93.3%	83.8%	77.1%	53.0%	26	12.6%	—	Full profileView
#7	API external/openai/gpt-4o Full profile	—	60.6%	86.6%	69.6%	74.3%	53.5%	—	31.9%	—	Full profileView
#8	Open moonshotai/Kimi-K2-Instruct Full profile	—	59.1%	65.4%	72.5%	80.4%	82.0%	—	12.1%	—	Full profileView

Dimensions:IQ—Reasoning & logicEQ—Social & emotionalAccuracy—Factual precisionCreativity—Creative outputBased—Direct & uncensored