Compare AI models

Specifications
Parameters
>1T106B
Context window
256k128k
API pricing
Input price
—$0.20
Output price
—$1.10
Cached input price
—$0.03
Cheapest input
—$0.13Novita
Cheapest output
—$0.85Novita
—$68.50Novita · bf16
Benchmarks
SWE-Bench Verified
75.3%57.6%
LiveCodeBench
85.9%70.7%
Humanity's Last Exam · text only, no tools
30.2%10.6%
GPQA Diamond
87.4%75%
MMLU-Pro
85.7%81.4%
Overview
CompanyQwenZ.ai
Release dateJan 26 2026Jul 28 2025
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen3-Max-Thinking leads GLM-4.5-Air on 5 of the 5 benchmarks they both report. Only GLM-4.5-Air has a verified first-party price for an available API: $0.20 per million input tokens and $1.10 per million output tokens. No currently available pay-as-you-go API rate is tracked for Qwen3-Max-Thinking. GLM-4.5-Air shipped 182 days before Qwen3-Max-Thinking, so benchmark comparisons should account for the intervening progress.

Qwen3-Max-Thinking has >1T parameters, while GLM-4.5-Air has 106B. Context windows are 256k (Qwen3-Max-Thinking) vs 128k (GLM-4.5-Air). Qwen3-Max-Thinking is closed, while GLM-4.5-Air is open weight.

On SWE-Bench Verified, Qwen3-Max-Thinking leads at 75.3% vs GLM-4.5-Air at 57.6%. On LiveCodeBench, Qwen3-Max-Thinking leads at 85.9% vs GLM-4.5-Air at 70.7%. On Humanity's Last Exam · text only, no tools, Qwen3-Max-Thinking leads at 30.2% vs GLM-4.5-Air at 10.6%. On GPQA Diamond, Qwen3-Max-Thinking leads at 87.4% vs GLM-4.5-Air at 75%. On MMLU-Pro, Qwen3-Max-Thinking leads at 85.7% vs GLM-4.5-Air at 81.4%.