Gemini 1.5 Flash 002: Complete Benchmark Performance Guide

google 1.5-flash-002 2024-09-24 Proprietary

In-depth analysis of Gemini 1.5 Flash 002 performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendorgoogle
Version1.5-flash-002
Release Date2024-09-24
Context Window1e+06 tokens
Input Modalitiestext, image, audio, video
Output Modalitiestext
LicenseProprietary
Documentationhttps://ai.google.dev/gemini-api/docs

Benchmark Performance

BenchmarkScoreUnitEvaluated AtNotesSource
MMLU81.1%2024-09-245-shotview
HUMANEVAL66.2pass@12024-09-24—view
GSM8K78.8%2024-09-240-shot CoTview
MATH40.5%2024-09-240-shot CoTview
BBH79.8%2024-09-243-shot CoTview
GPQA32%2024-09-240-shotview
IFEVAL77.2%2024-09-24prompt_strictview
ARC91.8%2024-09-24challengeview
MUSR59.8%2024-09-240-shotview
WINOGRANDE81.9%2024-09-240-shotview

Pricing

TierPriceCurrency
Input$0.075 / MtokUSD
Output$0.3 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://ai.google.dev/pricing · as of 2024-09-24

Compliance

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

Gemini 1.5 Flash 002

Modeloverzicht

Google Gemini 1.5 Flash 002 生产版本, 1M 上下文, 改进质量与稳定性。

Kernspecificaties

LeverancierVersieReleasedatumContextvensterInvoermodaliteitenUitvoermodaliteitenLicentie
Google1.5-flash-0022024-09-241000Ktext, image, audio, videotextProprietary

Benchmarkprestaties

BenchmarkScoreEenheidNotities
MMLU (Massive Multitask Language Understanding)81.1%5-shot
HumanEval66.2pass@1—
GSM8K (Grade School Math 8K)78.8%0-shot CoT
MATH40.5%0-shot CoT
BBH (BIG-Bench Hard)79.8%3-shot CoT
GPQA32.0%0-shot
IFEval77.2%prompt_strict
ARC91.8%challenge
MUSR59.8%0-shot
WinoGrande81.9%0-shot

Prijzen

InvoerUitvoerCache-lezenCache-schrijven
————

per miljoen tokens

Sterktes

  • MMLU score 81.1, strong knowledge reasoning.
  • 支持文本、图像、音频多模态输入。
  • 上下文窗口 1000K，支持长文本。

Zwaktes

  • 闭源专有模型，不支持自托管。

Gebruiksscenario’s

  • 长文档摘要
  • 视觉与图像理解

Referenties