Gemini 1.0 Pro: Complete Benchmark Performance Guide

google 1.0-pro 2023-12-06 Proprietary

In-depth analysis of Gemini 1.0 Pro performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendorgoogle
Version1.0-pro
Release Date2023-12-06
Context Window32768 tokens
Input Modalitiestext
Output Modalitiestext
LicenseProprietary
Documentationhttps://ai.google.dev/gemini-api/docs

Benchmark Performance

Benchmark Score Unit Evaluated At Notes Source
MMLU 84.5 % 2023-12-06 5-shot view
HUMANEVAL 76.4 pass@1 2023-12-06 — view
GSM8K 82.4 % 2023-12-06 0-shot CoT view
MATH 58.2 % 2023-12-06 0-shot CoT view
BBH 79.1 % 2023-12-06 3-shot CoT view
GPQA 47.7 % 2023-12-06 0-shot view
IFEVAL 75.1 % 2023-12-06 prompt_strict view
ARC 94.1 % 2023-12-06 challenge view
MUSR 57.8 % 2023-12-06 0-shot view
WINOGRANDE 84.5 % 2023-12-06 0-shot view

Pricing

Tier Price Currency
Input$0.5 / MtokUSD
Output$1.5 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://ai.google.dev/pricing · as of 2023-12-06

Compliance

  • Data Residency: US
  • SOC2: ✓
  • HIPAA: ✗
  • GDPR: ✓
  • ISO 27001: ✓

Gemini 1.0 Pro

Model Overview

Google Gemini 1.0 Pro 首代模型, 32K 上下文, 在 MMLU/GSM8K 上超越 GPT-3.5, 多模态能力初步集成。

Core Specifications

Vendor Version Release Date Context Window Input Modalities Output Modalities License
Google 1.0-pro 2023-12-06 32K text text Proprietary

Benchmark Performance

Benchmark Score Unit Notes
MMLU (Massive Multitask Language Understanding) 84.5 % 5-shot
HumanEval 76.4 pass@1 —
GSM8K (Grade School Math 8K) 82.4 % 0-shot CoT
MATH 58.2 % 0-shot CoT
BBH (BIG-Bench Hard) 79.1 % 3-shot CoT
GPQA 47.7 % 0-shot
IFEval 75.1 % prompt_strict
ARC 94.1 % challenge
MUSR 57.8 % 0-shot
WinoGrande 84.5 % 0-shot

Pricing

Input Output Cache Read Cache Write
— — — —

per million tokens

Strengths

  • MMLU score 84.5, strong knowledge reasoning.

Weaknesses

  • Proprietary, not self-hostable.

Use Cases

  • Code generation and debugging

References