You are here



Avalam: only the last level counts


Students were divided in groups of 2. Each group had to hand in one AI agent for the contest. Every technique was allowed as long as it was fair (e.g., they could not use CPU time while the opponent was playing). Each player got a time credit of 20 minutes per match.

We received 36 submissions that we divided randomly into 6 pools of 6 agents. All agents inside one pool played twice against all other agents in the same pool. For each match, the winner got 1 points. A draw game gave 0 point to the two opponents. In each pool, the two agents having the highest number of points were qualified. Then, the four best third place agents from each pool were also qualified. In case of ties, the slowest agent was eliminated (we summed the time the agent took to return each action).

The remaining 16 agents participated to a best-of-four playoff. For each match, the winner was the agent with the most points. In case of ties, the faster agent was qualified for the next level. The first place agents were playing against third place agents or second place agents. For each match, the winner went to the next level. A third place playoff was also played to determine the third place between the losers of the semi-final.


Position Students
1 Vincent Schellekens and Florimond Houssiaux
2 Cédric Vanden Bulcke and Charles Ninane

Vincent Schellekens, Florimond Houssiaux and Charles Ninane receiving their prize kindly sponsored by Case Départ

Match against a human

Vincent Sélenne is the editor of Avalam. He accepted to play two matches against the winning AI. It was two very exciting games.

And the winner was…

The AI twice! (It played once as first player then as second player).

The code

The code of the game has a GPLv2 license and is downloadable at the bottom of the page. Python 3.4-3.*.

Each AI agent lies in its own directory.

To start the game, go to the top directory and launch


assuming the two agents are located respectively at PATH_TO_AGENT1 and PATH_TO_AGENT2. You may also replace the path by the keyword human if you want to play against the computer. There are also other options, see python3 -h.

img_20151217_085320.jpg1015.61 KB
agents.zip594.67 KB
tournamenttraces.zip440.01 KB
avalamframework.zip306.25 KB