GLM-4 9B Chat: Complete Benchmark Performance Guide

other glm-4-9b-chat 2024-06-05 GLM-4 License

In-depth analysis of GLM-4 9B Chat performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendorother
Versionglm-4-9b-chat
Release Date2024-06-05
Context Window131072 tokens
Input Modalitiestext, image
Output Modalitiestext
LicenseGLM-4 License
Documentationhttps://huggingface.co/models

Benchmark Performance

Benchmark Score Unit Evaluated At Notes Source
MMLU 70.7 % 2024-06-05 5-shot view
HUMANEVAL 62.9 pass@1 2024-06-05 — view
GSM8K 58 % 2024-06-05 0-shot CoT view
MATH 26.6 % 2024-06-05 0-shot CoT view
BBH 67.2 % 2024-06-05 3-shot CoT view
GPQA 37.2 % 2024-06-05 0-shot view
IFEVAL 59.9 % 2024-06-05 prompt_strict view
ARC 88.6 % 2024-06-05 challenge view
MUSR 49.5 % 2024-06-05 0-shot view
WINOGRANDE 79.4 % 2024-06-05 0-shot view

Pricing

Tier Price Currency
Input$0.18 / MtokUSD
Output$0.18 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://huggingface.co/models · as of 2024-06-05

Compliance

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

GLM-4 9B Chat

Model Overview

清华智谱 GLM-4 9B Chat 开源对话模型, 131K 上下文, 9B 参数, 支持图像理解, 中英双语突出。

Core Specifications

Vendor Version Release Date Context Window Input Modalities Output Modalities License
Other glm-4-9b-chat 2024-06-05 131K text, image text GLM-4 License

Benchmark Performance

Benchmark Score Unit Notes
MMLU (Massive Multitask Language Understanding) 70.7 % 5-shot
HumanEval 62.9 pass@1 —
GSM8K (Grade School Math 8K) 58.0 % 0-shot CoT
MATH 26.6 % 0-shot CoT
BBH (BIG-Bench Hard) 67.2 % 3-shot CoT
GPQA 37.2 % 0-shot
IFEval 59.9 % prompt_strict
ARC 88.6 % challenge
MUSR 49.5 % 0-shot
WinoGrande 79.4 % 0-shot

Pricing

Input Output Cache Read Cache Write
— — — —

per million tokens

Strengths

  • Input Modalities: text, image, audio.

Weaknesses

  • Proprietary, not self-hostable.

Use Cases

  • Vision and image understanding

References