Humanity's Last Exam — cost vs performance

Best current row per canonical model in the group “accuracy · evaluator=Artificial Analysis” (453 models) against context window (tokens). 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 (10 models)

Frontier models 10

Pareto frontier
Model	accuracy on Humanity's Last Exam	Rank	context window (tokens)	Provider	Trust
Claude Fable 5.1Anthropic · Closed	59.1%	1	1M	—	Independent
Grok 4.6xAI · Closed	44.1%	15	500K	—	Independent
gpt-5.3-codexOpenAI · Closed	42.5%	21	400K	—	Independent
muse-sparkMeta AI · Closed	40.7%	29	262.1K	—	Independent
grok-build-0-1-06-16SpaceXAI · Closed	38.3%	40	256K	—	Independent
Claude Opus 4.5Anthropic · Closed	30.1%	74	200K	—	Independent
deepseek-v3-2-specialeDeepSeek · Open weights	28.7%	81	128K	—	Independent
qwen3-235b-a22b-instructAlibaba Group · Open weights	11.0%	184	32.8K	—	Independent
LFM2.5-1.2B-InstructLiquid AI · Restricted weights	6.72%	251	32K	—	Independent
olmo-2-7bAllen Institute for AI · Open weights	5.38%	291	4.1K	—	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.