GLM-4-Air

Released

GLM-4-Air is an AI model released by Z.ai on Wednesday, Jun 5 2024, the same day as GLM-4-9B. It has a 128k token context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider.. Benchmark results (shown below) cover HumanEval, IFEval, MATH, MMLU, and GSM8K.

Benchmarks

Coding

HumanEvalFunction synthesis — 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better.
75.7%
#41 of 74

Agentic & tool use

IFEvalAgentic tasks — following verifiable instructions in generated responses
75.2%
#9 of 11

Reasoning & science

MATHReasoning — accuracy on the original competition mathematics test set
57.9%
#22 of 50
MMLUGeneral knowledge — A 57-subject multiple-choice exam — history, law, medicine, maths — that was the standard measure of how much a model knows from 2020 until roughly 2024, when frontier scores crowded into the high 80s and labs moved on to harder tests. The scores here were published years apart under different testing setups, so read them as a historical record rather than a like-for-like ranking. Higher is better.
81.9%
#31 of 68
GSM8KGrade-school math — Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better.
90.9%
#11 of 39

Compare GLM-4-Air with

GLM-4-Air

About

Zhipu AI introduced GLM-4-Air at its June 2024 Open Day as a lower-cost and faster hosted model in the GLM-4 family. It supported a 128K-token context and targeted similar capabilities to the January GLM-4 flagship.

The GLM team's technical report identifies the release as the 0605 checkpoint. It describes upgrades to pretraining and alignment and reports 81.9% on MMLU, 90.9% on GSM8K and 75.2% on English IFEval under prompt-level strict scoring.

Frequently asked questions

GLM-4-Air was released by Z.ai on Wednesday, Jun 5 2024.

All Z.ai releases

24 tracked