Compare AI models

Specifications
Parameters
27B744B
Context window
262k—
API pricing
Input price
—$1.00
Output price
—$3.20
Cached input price
—$0.20
Cheapest input
$0.05Darkbloom$0.60GMICloud
Cheapest output
$1.875DeepInfra$1.92GMICloud
Benchmarks
BullshitBench v2
79%28%
Humanity's Last Exam · with tools
30.8%50.4%
GPQA Diamond
89.2%86%
Overview
CompanyQwenZ.ai
Release dateAug 14 2026Feb 12 2026
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen3.8-27B leads GLM-5 on 2 of the 3 benchmarks they both report (BullshitBench v2, Humanity's Last Exam, GPQA Diamond). Only GLM-5 has a verified first-party API price: $1.00 per million input tokens and $3.20 per million output tokens. No pay-as-you-go API rate is tracked for Qwen3.8-27B. GLM-5 shipped 183 days before Qwen3.8-27B, so benchmark comparisons should account for the intervening progress.

Qwen3.8-27B has 27B parameters, while GLM-5 has 744B.

On BullshitBench v2, Qwen3.8-27B leads at 79% vs GLM-5 at 28%. On Humanity's Last Exam · with tools, GLM-5 leads at 50.4% vs Qwen3.8-27B at 30.8%. On GPQA Diamond, Qwen3.8-27B leads at 89.2% vs GLM-5 at 86%.