Key Specifications

Vendormeta
Version3-70b
Release Date2024-04-18
Context Window8192 tokens
Input Modalitiestext
Output Modalitiestext
LicenseLlama 3 Community License
Documentationhttps://llama.meta.com/docs/

Benchmark Performance

Benchmark Score Unit Evaluated At Notes Source
MMLU 79.5 % 2024-04-18 5-shot view
HUMANEVAL 73 pass@1 2024-04-18 view
GSM8K 76.2 % 2024-04-18 0-shot CoT view
MATH 50.1 % 2024-04-18 0-shot CoT view
BBH 77.6 % 2024-04-18 3-shot CoT view
GPQA 38.2 % 2024-04-18 0-shot view
IFEVAL 72.2 % 2024-04-18 prompt_strict view
ARC 93.4 % 2024-04-18 challenge view
MUSR 51.2 % 2024-04-18 0-shot view
WINOGRANDE 79.2 % 2024-04-18 0-shot view

Pricing

Tier Price Currency
Input$0.9 / MtokUSD
Output$0.9 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://ai.meta.com/blog/ · as of 2024-04-18

Compliance

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

Llama 3 70B

Model Overview

Meta Llama 3 70B 中端开源模型, 8K 上下文, 在 MMLU/HumanEval 上接近 GPT-4, 开源旗舰之一。

Core Specifications

Vendor Version Release Date Context Window Input Modalities Output Modalities License
Meta 3-70b 2024-04-18 8K text text Llama 3 Community License

Benchmark Performance

Benchmark Score Unit Notes
MMLU (Massive Multitask Language Understanding) 79.5 % 5-shot
HumanEval 73.0 pass@1
GSM8K (Grade School Math 8K) 76.2 % 0-shot CoT
MATH 50.1 % 0-shot CoT
BBH (BIG-Bench Hard) 77.6 % 3-shot CoT
GPQA 38.2 % 0-shot
IFEval 72.2 % prompt_strict
ARC 93.4 % challenge
MUSR 51.2 % 0-shot
WinoGrande 79.2 % 0-shot

Pricing

Input Output Cache Read Cache Write

per million tokens

Strengths

  • Reliable general-purpose model.

Weaknesses

  • Proprietary, not self-hostable.
  • Context window 8K is limited.

Use Cases

  • Code generation and debugging

References