MMMU-Pro — cost vs performance

Best current row per canonical model in the group “accuracy · evaluator=Artificial Analysis” (157 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 (11 models)bubble = context window

Frontier models 11

Pareto frontier
Model	accuracy on MMMU-Pro	Rank	context window (tokens)	Provider	Trust
gpt-6-astraOpenAI · Closed	86.9%	1	1.05M	—	Independent
Gemini 3.8 FlashGoogle · Closed	85.6%	2	1.05M	—	Independent
Claude Opus 5Anthropic · Closed	84.7%	4	1M	—	Independent
muse-sparkMeta AI · Closed	80.5%	13	262.1K	—	Independent
apodex-1-1Apodex · Closed	79.2%	22	256K	—	Independent
Ling 3.0 Flash VLinclusionAI · Open weights	79.0%	24	131.1K	—	Independent
ernie-5-0-thinking-previewBaidu · Closed	64.6%	92	128K	—	Independent
step-3-vl-10bStepFun · Open weights	64.0%	96	65.5K	—	Independent
Molmo2-8BAllen Institute for AI · Open weights	37.5%	148	36.9K	—	Independent
LFM2.5-VL-1.6BLiquid AI · Open weights	26.5%	153	32K	—	Independent
molmo-7b-dAllen Institute for AI · Open weights	24.5%	156	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.