GPQA Diamond — cost vs performance

Best current row per canonical model in the group “accuracy · variant=Diamond · evaluator=Artificial Analysis” (440 models) against total parameters. 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 = context window

Frontier models 9

Pareto frontier
Model	accuracy on GPQA Diamond	Rank	total parameters	Provider	Trust
Kimi K3Moonshot AI · Open weights	93.5%	10	2.78T	—	Independent
MiniMax-M3MiniMax · Open weights	92.9%	14	427B	—	Independent
GLM 5.3 FlashZ.ai (Zhipu AI) · Open weights	91.2%	30	321.3B	—	Independent
Qwen3.8 27BQwen · Open weights	90.5%	37	27.8B	—	Independent
Qwen3.5-9BQwen · Open weights	80.6%	135	9.65B	—	Independent
Qwen3.5-4BQwen · Open weights	77.1%	162	4.66B	—	Independent
LFM2.5-2.6B (free)Liquid AI · Restricted weights	55.8%	290	2.7B	—	Independent
exaone-4-0-1-2bLG AI Research · Open weights	51.5%	306	1.28B	—	Independent
LFM2.5-1.2B-InstructLiquid AI · Restricted weights	32.6%	390	1.17B	—	Independent

Methodology. Points are the best current row per canonical model in comparability group 'accuracy · variant=Diamond · 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.