Nous Capybara 7B: Complete Benchmark Performance Guide

other nous-capybara-7b 2023-09-26 Apache 2.0

In-depth analysis of Nous Capybara 7B performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendorother
Versionnous-capybara-7b
Release Date2023-09-26
Context Window4096 tokens
Input Modalitiestext
Output Modalitiestext
LicenseApache 2.0
Documentationhttps://huggingface.co/models

Benchmark Performance

BenchmarkScoreUnitEvaluated AtNotesSource
MMLU64.4%2023-09-265-shotview
HUMANEVAL38.8pass@12023-09-26—view
GSM8K31%2023-09-260-shot CoTview
MATH19.7%2023-09-260-shot CoTview
BBH49.6%2023-09-263-shot CoTview
GPQA31%2023-09-260-shotview
IFEVAL44.2%2023-09-26prompt_strictview
ARC87%2023-09-26challengeview
MUSR42.5%2023-09-260-shotview
WINOGRANDE77.1%2023-09-260-shotview

Pricing

TierPriceCurrency
Input$0.18 / MtokUSD
Output$0.18 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://huggingface.co/models · as of 2023-09-26

Compliance

  • Data Residency: self-host
  • SOC2: ✗
  • HIPAA: ✗
  • GDPR: ✗
  • ISO 27001: ✗

Nous Capybara 7B

Modellöversikt

NousResearch Capybara 7B 对话模型, 4K 上下文, 基于 Llama 2 微调, 改进多轮对话能力。

Kärnspecifikationer

LeverantörVersionReleasedatumKontextfönsterInmatningsmodaliteterUtmatningsmodaliteterLicens
Othernous-capybara-7b2023-09-264KtexttextApache 2.0

Benchmarkprestanda

BenchmarkPoängEnhetAnteckningar
MMLU (Massive Multitask Language Understanding)64.4%5-shot
HumanEval38.8pass@1—
GSM8K (Grade School Math 8K)31.0%0-shot CoT
MATH19.7%0-shot CoT
BBH (BIG-Bench Hard)49.6%3-shot CoT
GPQA31.0%0-shot
IFEval44.2%prompt_strict
ARC87.0%challenge
MUSR42.5%0-shot
WinoGrande77.1%0-shot

Priser

InmatningUtmatningCacheläsningCacheskrivning
————

per miljon tokens

Styrkor

  • 可靠的通用模型。

Svagheter

  • HumanEval 38.8，代码能力较弱。
  • 闭源专有模型，不支持自托管。
  • 上下文窗口 4K 偏小。

Användningsområden

  • 通用对话与问答

Referenser