Claude 3 Haiku: Complete Benchmark Performance Guide

anthropic 3-haiku 2024-03-07 Proprietary

In-depth analysis of Claude 3 Haiku performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendoranthropic
Version3-haiku
Release Date2024-03-07
Context Window200000 tokens
Input Modalitiestext, image
Output Modalitiestext
LicenseProprietary
Documentationhttps://docs.anthropic.com/claude/docs

Benchmark Performance

BenchmarkScoreUnitEvaluated AtNotesSource
MMLU77.9%2024-03-075-shotview
HUMANEVAL69.4pass@12024-03-07—view
GSM8K82.2%2024-03-070-shot CoTview
MATH52.2%2024-03-070-shot CoTview
BBH75.7%2024-03-073-shot CoTview
GPQA44.4%2024-03-070-shotview
IFEVAL70.8%2024-03-07prompt_strictview
ARC94.3%2024-03-07challengeview
MUSR58.1%2024-03-070-shotview
WINOGRANDE84.8%2024-03-070-shotview

Pricing

TierPriceCurrency
Input$0.25 / MtokUSD
Output$1.25 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://www.anthropic.com/pricing · as of 2024-03-07

Compliance

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

Claude 3 Haiku

Modeloverzicht

Anthropic Claude 3 Haiku 经济型模型, 200K 上下文, 响应速度最快, 适合高吞吐量实时场景。

Kernspecificaties

LeverancierVersieReleasedatumContextvensterInvoermodaliteitenUitvoermodaliteitenLicentie
Anthropic3-haiku2024-03-07200Ktext, imagetextProprietary

Benchmarkprestaties

BenchmarkScoreEenheidNotities
MMLU (Massive Multitask Language Understanding)77.9%5-shot
HumanEval69.4pass@1—
GSM8K (Grade School Math 8K)82.2%0-shot CoT
MATH52.2%0-shot CoT
BBH (BIG-Bench Hard)75.7%3-shot CoT
GPQA44.4%0-shot
IFEval70.8%prompt_strict
ARC94.3%challenge
MUSR58.1%0-shot
WinoGrande84.8%0-shot

Prijzen

InvoerUitvoerCache-lezenCache-schrijven
————

per miljoen tokens

Sterktes

  • 可靠的通用模型。

Zwaktes

  • 闭源专有模型，不支持自托管。

Gebruiksscenario’s

  • 通用对话与问答

Referenties