Hermes 3 Llama 3.1 70B: Complete Benchmark Performance Guide

other hermes-3-llama-3-1-70b 2024-08-12 Llama 3.1 Community License

In-depth analysis of Hermes 3 Llama 3.1 70B performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendorother
Versionhermes-3-llama-3-1-70b
Release Date2024-08-12
Context Window131072 tokens
Input Modalitiestext
Output Modalitiestext
LicenseLlama 3.1 Community License
Documentationhttps://huggingface.co/models

Benchmark Performance

BenchmarkScoreUnitEvaluated AtNotesSource
MMLU76.7%2024-08-125-shotview
HUMANEVAL75.8pass@12024-08-12—view
GSM8K76.4%2024-08-120-shot CoTview
MATH54.4%2024-08-120-shot CoTview
BBH75.9%2024-08-123-shot CoTview
GPQA30.1%2024-08-120-shotview
IFEVAL73.8%2024-08-12prompt_strictview
ARC91.4%2024-08-12challengeview
MUSR55.4%2024-08-120-shotview
WINOGRANDE83.9%2024-08-120-shotview

Pricing

TierPriceCurrency
Input$0.8 / MtokUSD
Output$0.8 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://huggingface.co/models · as of 2024-08-12

Compliance

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

Hermes 3 Llama 3.1 70B

Model Overview

NousResearch Hermes 3 Llama 3.1 70B 微调模型, 131K 上下文, 改进指令遵循与中立性。

Core Specifications

VendorVersionRelease DateContext WindowInput ModalitiesOutput ModalitiesLicense
Otherhermes-3-llama-3-1-70b2024-08-12131KtexttextLlama 3.1 Community License

Benchmark Performance

BenchmarkScoreUnitNotes
MMLU (Massive Multitask Language Understanding)76.7%5-shot
HumanEval75.8pass@1—
GSM8K (Grade School Math 8K)76.4%0-shot CoT
MATH54.4%0-shot CoT
BBH (BIG-Bench Hard)75.9%3-shot CoT
GPQA30.1%0-shot
IFEval73.8%prompt_strict
ARC91.4%challenge
MUSR55.4%0-shot
WinoGrande83.9%0-shot

Pricing

InputOutputCache ReadCache Write
————

per million tokens

Strengths

  • Reliable general-purpose model.

Weaknesses

  • Proprietary, not self-hostable.

Use Cases

  • Code generation and debugging

References