This website uses cookies for authentication, support chat, and analytics. Only authentication cookies are turned on by default.
Don’t Enable
Enable All
Statistics
Templates
Feed
Public Projects
Log In
Animate this data
Bar chart race of DeepSeek-V3.2-Exp_thinking, gemini-4-argon_high, gpt-6-astra_max and 7 more, by GPQA Diamond.
Create it free
Takes a free account
DeepSeek-V3.2-Exp_thinking
Aider Polyglot
74 %
Cl bench
13 %
Cl bench Life
10 %
Critpt
3 %
Dtbench
88 %
Lmca
29 %
SciCode
39 %
WebDev Arena
1,367
WeirdML
39 %
Training Compute
4,180,000,000,000.00T FLOP
DeepSeek-V3.2-Exp_thinking benchmark results.
Aider Polyglot Monthly
Aider Polyglot score.
Aider Polyglot Yearly
Aider Polyglot score.
Cl bench Monthly
Cl bench score.
Cl bench Yearly
Cl bench score.
Cl bench Life Monthly
Cl bench Life score.
Cl bench Life Yearly
Cl bench Life score.
Critpt Monthly
Critpt score.
Critpt Yearly
Critpt score.
Dtbench Monthly
Dtbench score.
Dtbench Yearly
Dtbench score.
Lmca Monthly
Lmca score.
Lmca Yearly
Lmca score.
SciCode Monthly
SciCode score.
SciCode Yearly
SciCode score.
WebDev Arena Monthly
WebDev Arena score.
WebDev Arena Yearly
WebDev Arena score.
WeirdML Monthly
WeirdML score.
WeirdML Yearly
WeirdML score.
Data Source
Epoch AI, 'Capabilities & Benchmarking'. Published online at epoch.ai.
Related
DeepSeek-V3-0324
DeepSeek-V3.1
DeepSeek-V3.1-Terminus
DeepSeek-V3.1_thinking
DeepSeek-V3.2-Exp
DeepSeek-V3.2-Exp_unknown
DeepSeek-V3.2-Speciale
DeepSeek-V3-Base
deepseek-v4.1-flash-1017_high
deepseek-v4.1-flash-1017_low
View all in AI Benchmarks by Model
Updates
Tutorials
Language
English
Projects
Feed
Explore
Templates
Sign in
Log In
Statistics
Templates
Feed
Public Projects
Earn
Feature Requests
Blog
Theme
Language
English
Terms
Privacy
v3.65.1