Skip to content
AI Atlas

LiveBench Data Analysis — cost vs performance

Best current row per canonical model in the group “average score · variant=Data Analysis” (57 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.

open / restricted weights closed Pareto frontier (3 models)bubble = total parametersEstimated

Frontier models 3

Pareto frontier
Modelaverage score on LiveBench Data AnalysisRankESTIMATED memory at 4-bit, 8K context (GB)ProviderTrust
DeepSeek V4 Flash Vision ExpDeepSeek · Open weights79.5%8175.7 GBest.Official board
DeepSeek V4 Flash (0731)DeepSeek · Open weights79.3%9175.4 GBest.Official board
Qwen3.8 27BQwen · Open weights76.6%2416.5 GBest.Official board

Methodology. Points are the best current row per canonical model in comparability group 'average score · variant=Data 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.