New: GPT-6.2 benchmark report is here.Read the report ↗

Best LLMs on PostTrainBench Lite

PostTrainBench Lite

Improves an already-trained language model through further training within a fixed compute and time budget.

Updated: Data as of:
Limited reportingNot saturated
Official benchmark
PostTrainBench Lite results. Each row identifies its benchmark version and evaluation track. Scores from different versions are not directly comparable. Best effort from the latest stored report date per model and version.
Model / creatorVersionReported score
GPT-6.2 AstraOpenAI · ProprietaryGPT-6.2 system card74.79%
GPT-6.2 SolOpenAI · ProprietaryGPT-6.2 system card44.89%
View all 2 published reports