DeepSeek-V4-Pro-0813vsGLM-4.5

DeepSeek-V4-Pro-0813
GLM-4.5
Specifications
Parameters
355B
Context window
128k
API pricing
Input price
$0.66$0.60
Output price
$1.98$2.20
Cached input price
$0.022$0.11
Cheapest input
$0.96Ionstream
Cheapest output
$2.60DeepInfra
Benchmarks
BullshitBench v2
35%8%
SWE-Bench Verified
80.6%64.2%
Benchmarks
Terminal-Bench 2.1
87.9%
Toolathlon-Verified
74.1%
BrowseComp
83.4%
CyberGym
83.3%
Humanity's Last Exam · no tools
42.7%
Humanity's Last Exam · with tools
60%
GPQA Diamond
79.1%
AutomationBench
31.8%
Overview
CompanyDeepSeekZ.ai
Release dateAug 13 2026Jul 28 2025
AccessProprietaryOpen Weight

Other comparisons

DeepSeek-V4-Pro-0813vsClaude Fable 5.1GLM-4.5vsClaude Fable 5.1DeepSeek-V4-Pro-0813vsGPT-6 AstraGLM-4.5vsGPT-6 AstraDeepSeek-V4-Pro-0813vsGemini 3.8 FlashGLM-4.5vsGemini 3.8 FlashDeepSeek-V4-Pro-0813vsMuse Spark 1.3GLM-4.5vsMuse Spark 1.3DeepSeek-V4-Pro-0813vsGrok 4.6GLM-4.5vsGrok 4.6DeepSeek-V4-Pro-0813vsMistral Medium 3.5GLM-4.5vsMistral Medium 3.5

Frequently asked questions

DeepSeek-V4-Pro-0813 leads GLM-4.5 on 2 of the 2 benchmarks they both report (BullshitBench v2, SWE-Bench Verified). GLM-4.5 is cheaper on input: $0.60 vs $0.66 per million tokens. DeepSeek-V4-Pro-0813 is cheaper on output: $1.98 vs $2.20 per million tokens. Figures are base-tier rates. GLM-4.5 shipped 381 days before DeepSeek-V4-Pro-0813, so benchmark comparisons should account for the intervening progress.

DeepSeek-V4-Pro-0813 is proprietary, while GLM-4.5 is open weight.

On BullshitBench v2, DeepSeek-V4-Pro-0813 leads at 35% vs GLM-4.5 at 8%. On SWE-Bench Verified, DeepSeek-V4-Pro-0813 leads at 80.6% vs GLM-4.5 at 64.2%.