GLM-4 Flash: Complete Benchmark Performance Guide

other glm-4-flash 2024-08-12 Proprietary

In-depth analysis of GLM-4 Flash performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendorother
Versionglm-4-flash
Release Date2024-08-12
Context Window131072 tokens
Input Modalitiestext
Output Modalitiestext
LicenseProprietary
Documentationhttps://huggingface.co/models

Benchmark Performance

Benchmark Score Unit Evaluated At Notes Source
MMLU 80 % 2024-08-12 5-shot view
HUMANEVAL 67.6 pass@1 2024-08-12 — view
GSM8K 78.4 % 2024-08-12 0-shot CoT view
MATH 39.4 % 2024-08-12 0-shot CoT view
BBH 70.5 % 2024-08-12 3-shot CoT view
GPQA 31.3 % 2024-08-12 0-shot view
IFEVAL 65.9 % 2024-08-12 prompt_strict view
ARC 94.2 % 2024-08-12 challenge view
MUSR 56 % 2024-08-12 0-shot view
WINOGRANDE 81.1 % 2024-08-12 0-shot view

Pricing

Tier Price Currency
Input$0 / MtokUSD
Output$0 / 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: ✗

GLM-4 Flash

Model Overview

清华智谱 GLM-4 Flash 免费模型, 131K 上下文, 速度最快, 适合高并发实时场景。

Core Specifications

Vendor Version Release Date Context Window Input Modalities Output Modalities License
Other glm-4-flash 2024-08-12 131K text text Proprietary

Benchmark Performance

Benchmark Score Unit Notes
MMLU (Massive Multitask Language Understanding) 80.0 % 5-shot
HumanEval 67.6 pass@1 —
GSM8K (Grade School Math 8K) 78.4 % 0-shot CoT
MATH 39.4 % 0-shot CoT
BBH (BIG-Bench Hard) 70.5 % 3-shot CoT
GPQA 31.3 % 0-shot
IFEval 65.9 % prompt_strict
ARC 94.2 % challenge
MUSR 56.0 % 0-shot
WinoGrande 81.1 % 0-shot

Pricing

Input Output Cache Read Cache Write
— — — —

per million tokens

Strengths

  • MMLU score 80.0, strong knowledge reasoning.

Weaknesses

  • Proprietary, not self-hostable.

Use Cases

  • General chat and Q&A

References