Underground 2 Revived
Play Need for Speed: Underground 2 (2004) in Chrome and Firefox with WebGPU graphics, Web Audio, up to 120 fps widescreen, controller support, free roam and Online races with friends. Fan project, not
LLM agents play chess through tools, against each other and against you. Every request, reasoning trace, tool call, and taunt is stored, replayable, and one click from the leaderboard number it produc
GamesPython · TypeScriptGitHub @ahmedsaedOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
LLM agents playing chess — against each other, and against you.
Taken from the repository README
Loading the full README. The saved excerpt is shown below.
LLM agents playing chess — against each other, and against you.
1. A benchmark. Chess is a clean, adversarial, long-horizon agentic task. Models must use tools,
maintain state across dozens of turns, reason about consequences, and never emit an invalid
action. Chessmark measures how well they do it — legality, strength, cost, latency, and
consistency — with everything recorded.
2. A show. Reasoning streams live next to the board, move by move, as the model produces it.
You can sit down and play one yourself.Read the full README ↗The full README is unavailable right now. Read it at the original link. The saved excerpt is shown below.
LLM agents playing chess — against each other, and against you.
1. A benchmark. Chess is a clean, adversarial, long-horizon agentic task. Models must use tools,
maintain state across dozens of turns, reason about consequences, and never emit an invalid
action. Chessmark measures how well they do it — legality, strength, cost, latency, and
consistency — with everything recorded.
2. A show. Reasoning streams live next to the board, move by move, as the model produces it.
You can sit down and play one yourself.Read the full README ↗