← All models
Compare

DeepSeek-V3.2 vs GLM-5.1

Add, remove, or swap models to compare them side by side.

DeepSeek-V3.2GLM-5.1
DeepSeek-V3.2DeepSeek · 1 in the newsGLM-5.1Z.ai (Zhipu AI) · 5 in the news
Scores
Intelligence (ECI)146150
Coding2769
Math3663
Reasoning & Knowledge2648
Agentic & Tools29
Specifications
DeveloperDeepSeekZ.ai (Zhipu AI)
FamilyDeepSeek
ReleasedDec 1, 2025Apr 7, 2026
Parameters754B
AvailabilityOpen weights (unrestricted)
Context window131K203K
Price — $/M input$0.23$0.98
Price — $/M output$0.34$3.08
Inputstexttext
Outputstexttext
Benchmarks
AIME 2024/202588%92%
APEX7%
ARC-AGI57%
ARC-AGI-24%
FrontierMath22%33%
FrontierMath Tier 42%13%
GPQA Diamond83%85%
SimpleBench53%59%
SimpleQA Verified28%37%
Terminal-Bench40%
WebDev Arena12861534
WeirdML47%57%
SWE-bench Verified74%

Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.

Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.