Wordle Arena
Qwen3-1.7B playing Wordle before and after SFT and GRPO
None defined yet.
Qwen3-1.7B playing Wordle before and after SFT and GRPO
Play a Wordleโstyle guessing game with text input
Step through an echo environment with custom actions
Interact with a browser environment using text actions
Run Python code snippets in an interactive coding environment
Show interactive visualizations of tracked data
Single-tool HF Sandbox-backed coding environment
Step through a chat environment using token IDs
Interact with an OpenEnv agentic environment
Play OpenSpiel games via an interactive web interface
Control Atari games via web interface
Play a textโdriven Sudoku game
Run and control Python REPL agents in a web interface
Control traffic lights in a SUMO simulation via step actions
Interact with a Python REPL agent environment