Gemini 1.0 Flash: Complete Benchmark Performance Guide

google 1.0-flash 2024-02-15 Proprietary

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

Key Specifications

Vendorgoogle
Version1.0-flash
Release Date2024-02-15
Context Window32768 tokens
Input Modalitiestext, image
Output Modalitiestext
LicenseProprietary
Documentationhttps://ai.google.dev/gemini-api/docs

Benchmark Performance

Benchmark Score Unit Evaluated At Notes Source
MMLU 78.7 % 2024-02-15 5-shot view
HUMANEVAL 65.7 pass@1 2024-02-15 — view
GSM8K 86.2 % 2024-02-15 0-shot CoT view
MATH 39.9 % 2024-02-15 0-shot CoT view
BBH 74.8 % 2024-02-15 3-shot CoT view
GPQA 37.7 % 2024-02-15 0-shot view
IFEVAL 71.8 % 2024-02-15 prompt_strict view
ARC 92.6 % 2024-02-15 challenge view
MUSR 55.9 % 2024-02-15 0-shot view
WINOGRANDE 80.1 % 2024-02-15 0-shot view

Pricing

Tier Price Currency
Input$0.075 / MtokUSD
Output$0.3 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://ai.google.dev/pricing · as of 2024-02-15

Compliance

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

Gemini 1.0 Flash

Model Overview

Google Gemini 1.0 Flash 经济型, 32K 上下文, 速度快成本低, 适合实时多模态应用。

Core Specifications

Vendor Version Release Date Context Window Input Modalities Output Modalities License
Google 1.0-flash 2024-02-15 32K text, image text Proprietary

Benchmark Performance

Benchmark Score Unit Notes
MMLU (Massive Multitask Language Understanding) 78.7 % 5-shot
HumanEval 65.7 pass@1 —
GSM8K (Grade School Math 8K) 86.2 % 0-shot CoT
MATH 39.9 % 0-shot CoT
BBH (BIG-Bench Hard) 74.8 % 3-shot CoT
GPQA 37.7 % 0-shot
IFEval 71.8 % prompt_strict
ARC 92.6 % challenge
MUSR 55.9 % 0-shot
WinoGrande 80.1 % 0-shot

Pricing

Input Output Cache Read Cache Write
— — — —

per million tokens

Strengths

  • GSM8K 86.2, robust math reasoning.

Weaknesses

  • Proprietary, not self-hostable.

Use Cases

  • General chat and Q&A

References