The Home Page of UT Austin's TacTex Agent


Welcome to the home of the TacTex agent, the winner of three different Trading Agent Competitions (TAC).

In 2013-2015 TacTex was a winner and a top performer in the Power Trading Agent Competition (Power TAC).

This page focuses on TacTex's participation in Power TAC.

In the past, TacTex was the winner of both the TAC Ad Auctions (TAC AA) competition and the TAC Supply Chain Management (TAC SCM) competition. See links on the left for more details about these past competitions.


The TacTex team consists of:
TacTex is a project of the Learning Agents Research Group in the Department of Computer Science at the University of Texas at Austin.

Game Overview

Structure of the Power TAC Simulation Environment

Broker interactions with the simulation environment (source:

More details are in the Power TAC game specification ( A short version of the game description is in chapter 2 of Daniel Urieli's Ph.D dissertation Autonomous Trading in Modern Electricity Markets.

TacTex's Binaries:

The TacTex agent binaries that played in the Power TAC finals can be found here:
TacTex 2013
TacTex 2014
TacTex 2015
Note: Each of these binaries works with a corresponding version of the Power TAC server that was used in the finals in the same year. For details on getting and running these server versions see

TacTex's Source Code:

TacTex's complete source code is now in GitHub. See:


Ph.D. Dissertation:

Autonomous Trading in Modern Electricity Markets
Daniel Urieli
Ph.D. Dissertation, The University of Texas at Austin, Austin, Texas, USA, 2015.

Conference Papers:

An MDP-Based Winning Approach to Autonomous Power Trading: Formalization and Empirical Analysis
Daniel Urieli, Peter Stone
In Proc. of the 15th International Conference on Autonomous Agents and Multiagent Systems, 2016 (AAMAS-16).

Autonomous Electricity Trading using Time-Of-Use Tariffs in a Competitive Market
Daniel Urieli, Peter Stone
In Proc. of the 30th Conference on Artificial Intelligence, 2016 (AAAI-16).

TacTex'13: A Champion Adaptive Power Trading Agent.
Daniel Urieli, Peter Stone
In Proc. of the 28th Conference on Artificial Intelligence, 2014 (AAAI-14).