Gemini 1.0 Pro (001): benchmark results

Provider: Google. Released 2024-02-01. Access: API.

Unified ELO 1448 ± 1, rank #1065 of 1553 rated models, from 62 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
HELM MMLU - Computer Security84EM78.2
HELM MMLU - College Mathematics51EM66.7
HELM MMLU - Jurisprudence86.11EM57.1
MMLU70Accuracy (%)56.3
HELM MMLU - Sociology90.05EM55.1
HELM MMLU - Security Studies80.41EM50
HELM MMLU - International Law87.6EM47.4
HELM MMLU - World Religions85.96EM47.4
HELM MMLU - Global Facts49EM46.2
HELM MMLU - Human Aging77.58EM44.2
HELM MMLU - High School Computer Science79EM42.9
HELM MMLU - High School Biology86.13EM40.4

Interactive version: theaggregate.ai/model?slug=gemini-1-0-pro-001 · How It Works · Data refreshed daily, snapshot 2026-09-08.