Phi-3.5 Mini: Complete Benchmark Performance Guide

other phi-3-5-mini 2024-08-16 MIT

In-depth analysis of Phi-3.5 Mini performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendorother
Versionphi-3-5-mini
Release Date2024-08-16
Context Window128000 tokens
Input Modalitiestext
Output Modalitiestext
LicenseMIT
Documentationhttps://huggingface.co/models

Benchmark Performance

Benchmark Score Unit Evaluated At Notes Source
MMLU 55.7 % 2024-08-16 5-shot view
HUMANEVAL 44 pass@1 2024-08-16 — view
GSM8K 22.1 % 2024-08-16 0-shot CoT view
MATH 9.6 % 2024-08-16 0-shot CoT view
BBH 47.5 % 2024-08-16 3-shot CoT view
GPQA 24.4 % 2024-08-16 0-shot view
IFEVAL 48.7 % 2024-08-16 prompt_strict view
ARC 81.5 % 2024-08-16 challenge view
MUSR 31.2 % 2024-08-16 0-shot view
WINOGRANDE 68.6 % 2024-08-16 0-shot view

Pricing

Tier Price Currency
Input$0.1 / MtokUSD
Output$0.1 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://huggingface.co/models · as of 2024-08-16

Compliance

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

Phi-3.5 Mini

Model Overview

Microsoft Phi-3.5 Mini 3.8B 模型, 128K 上下文, 改进多语言与长上下文能力, 适合边缘部署。

Core Specifications

Vendor Version Release Date Context Window Input Modalities Output Modalities License
Other phi-3-5-mini 2024-08-16 128K text text MIT

Benchmark Performance

Benchmark Score Unit Notes
MMLU (Massive Multitask Language Understanding) 55.7 % 5-shot
HumanEval 44.0 pass@1 —
GSM8K (Grade School Math 8K) 22.1 % 0-shot CoT
MATH 9.6 % 0-shot CoT
BBH (BIG-Bench Hard) 47.5 % 3-shot CoT
GPQA 24.4 % 0-shot
IFEval 48.7 % prompt_strict
ARC 81.5 % challenge
MUSR 31.2 % 0-shot
WinoGrande 68.6 % 0-shot

Pricing

Input Output Cache Read Cache Write
— — — —

per million tokens

Strengths

  • Context window 128K.

Weaknesses

  • MMLU 55.7, weak knowledge reasoning.
  • HumanEval 44.0, coding weak.
  • Proprietary, not self-hostable.

Use Cases

  • Long document summarization

References