Qwen3.8-MaxvsGLM-5

Qwen3.8-Max
GLM-5
Specifications
Parameters
2.4T744B
Context window
1M
API pricing
Input price
$1.00
Output price
$3.20
Cached input price
$0.20
Cheapest input
$2.00Alibaba$0.60GMICloud
Cheapest output
$6.00Alibaba$1.92GMICloud
Benchmarks
BullshitBench v2
95%28%
Humanity's Last Exam · with tools
43.6%50.4%
Benchmarks
SWE-Bench Pro
67.7%
SWE-Bench Verified
77.8%
SWE-Bench Multilingual
73.3%
DeepSWE 1.1
56.6%
PaperBench
93%
NL2Repo-Bench
55.9%
QwenSWEBench V2
55.1%
Terminal-Bench 3.0
11.3%
Terminal-Bench 2.1
86.6%
Terminal-Bench 2.0
56.2%
JobBench
53.4%
CoWorkBench
74.8%
Toolathlon-Verified
72.5%
BrowseComp
75.9%
GPQA Diamond
86%
OSWorld-Verified
86.1%
CharXiv Reasoning
88.4%
BabyVision
82%
MMMU-Pro
82.3%
Overview
CompanyQwenZ.ai
Release dateAug 3 2026Feb 12 2026
AccessProprietaryOpen Weight

Other comparisons

Qwen3.8-MaxvsClaude Fable 5.1GLM-5vsClaude Fable 5.1Qwen3.8-MaxvsGPT-6 AstraGLM-5vsGPT-6 AstraQwen3.8-MaxvsGemini 3.8 FlashGLM-5vsGemini 3.8 FlashQwen3.8-MaxvsMuse Spark 1.3GLM-5vsMuse Spark 1.3Qwen3.8-MaxvsGrok 4.6GLM-5vsGrok 4.6Qwen3.8-MaxvsDeepSeek-V4.1-FlashGLM-5vsDeepSeek-V4.1-Flash

Frequently asked questions

Qwen3.8-Max and GLM-5 are evenly matched across the 2 benchmarks they both report (BullshitBench v2, Humanity's Last Exam). 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-Max. GLM-5 shipped 172 days before Qwen3.8-Max, so benchmark comparisons should account for the intervening progress.

Qwen3.8-Max has 2.4T parameters, while GLM-5 has 744B. Qwen3.8-Max is proprietary, while GLM-5 is open weight.

On BullshitBench v2, Qwen3.8-Max leads at 95% vs GLM-5 at 28%. On Humanity's Last Exam · with tools, GLM-5 leads at 50.4% vs Qwen3.8-Max at 43.6%.