Command R+ (08-2024): Complete Benchmark Performance Guide

cohere r-plus-08-2024 2024-08-13 CC-BY-NC-4.0

In-depth analysis of Command R+ (08-2024) performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendorcohere
Versionr-plus-08-2024
Release Date2024-08-13
Context Window128000 tokens
Input Modalitiestext
Output Modalitiestext
LicenseCC-BY-NC-4.0
Documentationhttps://docs.cohere.com/docs

Benchmark Performance

BenchmarkScoreUnitEvaluated AtNotesSource
MMLU82.3%2024-08-135-shotview
HUMANEVAL75pass@12024-08-13—view
GSM8K80.8%2024-08-130-shot CoTview
MATH43.3%2024-08-130-shot CoTview
BBH82.1%2024-08-133-shot CoTview
GPQA38.7%2024-08-130-shotview
IFEVAL72.1%2024-08-13prompt_strictview
ARC93.4%2024-08-13challengeview
MUSR51.1%2024-08-130-shotview
WINOGRANDE83.4%2024-08-130-shotview

Pricing

TierPriceCurrency
Input$2.5 / MtokUSD
Output$10 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://cohere.com/pricing · as of 2024-08-13

Compliance

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

Command R+ (08-2024)

모델 개요

Cohere Command R+ 08-2024 版本, 128K 上下文, 改进代码生成与多语言能力, 性能接近 GPT-4。

핵심 사양

공급업체버전출시일컨텍스트 창입력 모달리티출력 모달리티라이선스
Coherer-plus-08-20242024-08-13128KtexttextCC-BY-NC-4.0

벤치마크 성능

벤치마크점수단위비고
MMLU (Massive Multitask Language Understanding)82.3%5-shot
HumanEval75.0pass@1—
GSM8K (Grade School Math 8K)80.8%0-shot CoT
MATH43.3%0-shot CoT
BBH (BIG-Bench Hard)82.1%3-shot CoT
GPQA38.7%0-shot
IFEval72.1%prompt_strict
ARC93.4%challenge
MUSR51.1%0-shot
WinoGrande83.4%0-shot

가격

입력출력캐시 읽기캐시 쓰기
————

백만 토큰당

강점

  • MMLU score 82.3, strong knowledge reasoning.
  • 企业级合规认证。

약점

  • 闭源专有模型，不支持自托管。

사용 사례

  • 代码生成与调试
  • 企业客户支持

참고문헌