Skip to content

Calix-LDanKS

Repository stars
831Star
All-time
#46,272 all time
Fork
35 Fork
Repository created
since 2026
Website
View source

RL‐Empowered Small‐Scale Competitive Guandan Agent

TypeAgentsPurposeMiscellaneousGamingLanguagePythonLicenseApache-2.0
Commercial use OKApache-2.0; keep the license notice
Recently updatedLast push 21 days ago

About

DanKS is a small-scale competitive Guandan AI that combines structured retrieval with PPO learning to make better decisions across the whole hand.

It suits people studying agents, training policies, or building a local table. The project also offers two generations of open-source web service that run locally and can connect to custom models.

Summarized by AI from the READMEREADME

As of 2026-11-12, Calix-L/DanKS has 831 GitHub stars and 35 forks. Written mainly in Python; licensed under Apache-2.0; last pushed on 2026-10-31.Record updated:

Raw observations · 7
Historical observations
Measured atRepository starsSource
2026-11-12 05:44 UTC831Source
2026-11-11 05:54 UTC831Source
2026-11-10 04:45 UTC831Source
2026-11-09 04:12 UTC831Source
2026-11-08 04:12 UTC831Source
2026-11-07 20:28 UTC831Source
2026-11-06 12:38 UTC832Source

First collected 2026-11-06Measured at 2026-11-12 05:44 UTCSourceLicense source

Compiled from public project information; not independently tested. · Repository Stars, not unique users · License tiers are derived from metadata. Check obligations for your use; unknown does not mean prohibited.