Key Specifications

Vendormeta
Version3.1-70b
Release Date2024-07-23
Context Window128000 tokens
Input Modalitiestext
Output Modalitiestext
LicenseLlama 3 Community License
Documentationhttps://llama.meta.com/docs/

Benchmark Performance

BenchmarkScoreUnitEvaluated AtNotesSource
MMLU75.6%2024-07-235-shotview
HUMANEVAL79.7pass@12024-07-23view
GSM8K78.8%2024-07-230-shot CoTview
MATH38.5%2024-07-230-shot CoTview
BBH70.2%2024-07-233-shot CoTview
GPQA40%2024-07-230-shotview
IFEVAL73.7%2024-07-23prompt_strictview
ARC92.3%2024-07-23challengeview
MUSR48.1%2024-07-230-shotview
WINOGRANDE81%2024-07-230-shotview

Pricing

TierPriceCurrency
Input$0.9 / MtokUSD
Output$0.9 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://ai.meta.com/blog/ · as of 2024-07-23

Compliance

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

Llama 3.1 70B

Ikhtisar Model

Meta Llama 3.1 70B 中端开源模型, 128K 上下文, 性能接近 GPT-4, 推理与编码能力突出。

Spesifikasi Inti

VendorVersiTanggal RilisJendela KonteksModalitas InputModalitas OutputLisensi
Meta3.1-70b2024-07-23128KtexttextLlama 3 Community License

Kinerja Benchmark

BenchmarkSkorSatuanCatatan
MMLU (Massive Multitask Language Understanding)75.6%5-shot
HumanEval79.7pass@1
GSM8K (Grade School Math 8K)78.8%0-shot CoT
MATH38.5%0-shot CoT
BBH (BIG-Bench Hard)70.2%3-shot CoT
GPQA40.0%0-shot
IFEval73.7%prompt_strict
ARC92.3%challenge
MUSR48.1%0-shot
WinoGrande81.0%0-shot

Harga

InputOutputBaca CacheTulis Cache

per juta token

Kelebihan

  • 可靠的通用模型。

Kekurangan

  • 闭源专有模型,不支持自托管。

Kasus Penggunaan

  • 代码生成与调试

Referensi