Humanity's Last Exam — cost vs performance

Best current row per canonical model in the group “accuracy · evaluator=Artificial Analysis” (453 models) against ESTIMATED memory at 4-bit, 8K context (GB). The dashed line is the Pareto frontier: no model is both better and cheaper than a point on it.

← Leaderboard

XOutput price Input price Parameters Context Memory (est.)Scalelog linearBubblecontext params none

open / restricted weights closed Pareto frontier (9 models)bubble = total parametersEstimated

Frontier models 9

Pareto frontier
Model	accuracy on Humanity's Last Exam	Rank	ESTIMATED memory at 4-bit, 8K context (GB)	Provider	Trust
Kimi K3Moonshot AI · Open weights	46.9%	11	1.6K GBest.	—	Independent
GLM 5.3Z.ai (Zhipu AI) · Open weights	42.3%	24	433.7 GBest.	—	Independent
GLM 5.3 FlashZ.ai (Zhipu AI) · Open weights	39.9%	33	185.3 GBest.	—	Independent
Qwen3.8 27BQwen · Open weights	33.9%	59	16.5 GBest.	—	Independent
Gemma 4 26B A4BGoogle · Open weights	19.3%	127	15.3 GBest.	—	Independent
Qwen3.5-9BQwen · Open weights	14.9%	149	6.1 GBest.	—	Independent
Qwen3.5-4BQwen · Open weights	9.92%	205	3.2 GBest.	—	Independent
LFM2.5-1.2B-InstructLiquid AI · Restricted weights	6.72%	251	1.2 GBest.	—	Independent
Qwen3-0.6BQwen · Open weights	5.64%	280	0.9 GBest.	—	Independent

Methodology. Points are the best current row per canonical model in comparability group 'accuracy · evaluator=Artificial Analysis'. Price = cheapest current offer across providers (the provider shown). Pareto frontier maximises the score and minimises x; exact ties are all kept. memory_estimate is an estimate (see /methodology). Only the selected comparability group is plotted; points under other configurations are not mixed in. Price = cheapest current offer across providers at the time of the last crawl. Nothing is estimated except where marked.