MMLU (5-shot)Knowledge
85.9/ 100
Verified
Last Verified: Unknown DateGoogle AI Blog
Massive Multitask Language Understanding covers 57 subjects across STEM, the humanities, social sciences, and more.
Beta version: *Information might not be fully accurate. Please report any discrepancies.
Latest Data
Unknown
Context Window
2.0M
tokens
Input Cost
$3.50
per 1M tokens
Output Cost
$10.50
per 1M tokens
Parameters
MoE
model footprint
Performance Analysis // Verified Benchmarks
Massive Multitask Language Understanding covers 57 subjects across STEM, the humanities, social sciences, and more.
Challenging competition mathematics problems (AIME/IMO level).
Functional correctness of synthesized programs from docstrings.
Multi-discipline Multimodal Understanding and Reasoning.
Chatbot Arena ELO score. Crowd-sourced human preference ranking.
Graduate-Level Google-Proof Q&A Benchmark.