Key Specifications

Vendorother
Versionyi-34b
Release Date2023-11-05
Context Window4096 tokens
Input Modalitiestext
Output Modalitiestext
LicenseApache 2.0
Documentationhttps://huggingface.co/models

Benchmark Performance

Benchmark Score Unit Evaluated At Notes Source
MMLU 55.5 % 2023-11-05 5-shot view
HUMANEVAL 35.8 pass@1 2023-11-05 view
GSM8K 48.3 % 2023-11-05 0-shot CoT view
MATH 20.5 % 2023-11-05 0-shot CoT view
BBH 48.8 % 2023-11-05 3-shot CoT view
GPQA 20.6 % 2023-11-05 0-shot view
IFEVAL 48.2 % 2023-11-05 prompt_strict view
ARC 87.2 % 2023-11-05 challenge view
MUSR 36.8 % 2023-11-05 0-shot view
WINOGRANDE 71.5 % 2023-11-05 0-shot view

Pricing

Tier Price Currency
Input$0.4 / MtokUSD
Output$0.4 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://huggingface.co/models · as of 2023-11-05

Compliance

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

Yi 34B

Model Overview

01.AI Yi 34B 首代开源模型, 4K 上下文, 340 亿参数, 在开源模型中曾排名第一。

Core Specifications

Vendor Version Release Date Context Window Input Modalities Output Modalities License
Other yi-34b 2023-11-05 4K text text Apache 2.0

Benchmark Performance

Benchmark Score Unit Notes
MMLU (Massive Multitask Language Understanding) 55.5 % 5-shot
HumanEval 35.8 pass@1
GSM8K (Grade School Math 8K) 48.3 % 0-shot CoT
MATH 20.5 % 0-shot CoT
BBH (BIG-Bench Hard) 48.8 % 3-shot CoT
GPQA 20.6 % 0-shot
IFEval 48.2 % prompt_strict
ARC 87.2 % challenge
MUSR 36.8 % 0-shot
WinoGrande 71.5 % 0-shot

Pricing

Input Output Cache Read Cache Write

per million tokens

Strengths

  • Reliable general-purpose model.

Weaknesses

  • MMLU 55.5, weak knowledge reasoning.
  • HumanEval 35.8, coding weak.
  • Proprietary, not self-hostable.
  • Context window 4K is limited.

Use Cases

  • General chat and Q&A

References