
Badges
Activity
Challenge Categories
Challenges Entered
Build an LLM agent for five real-world games
Latest submissions
See All| graded | 304675 | ||
| graded | 304644 | ||
| graded | 304639 |
Machine Learning for detection of early onset of Alzheimers
Latest submissions
| Participant | Rating |
|---|
| Participant | Rating |
|---|
-
TheSquirrel Orak Game Agent Challenge 2025View
-
TheSquirrel Global Chess Challenge 2025View
Global Chess Challenge 2025
Orak Game Agent Challenge
Pokemon map is broken?
YesterdayThis submission, for instance: AIcrowd | Orak Game Agent Challenge 2025 | Submissions #304644
The game clearly started, the menu and initial dialog were passed. The location is RedsHouse1f. From what I can see, one of the problems might be that in the โmap_names.jsonโ it is โRedsHouse1Fโ, not โRedsHouse1fโ (with capital F).
The bigger problem in the code is the fact that โpokemon_tools.pyโ or โpokemon_tools_mcp.pyโ are not used anywhere at all. Including the โprocess_state_toolโ function that is supposed to replace โ?โ when the tile is explored. Another side effect of that is that advanced tools (such as โmove_toโ) are not callable.
Pokemon map is broken?
2 days agoIt seems that after the recent evals update, the map is broken both in local and remote modes. Most of the cells are โ?โ. I think before the update, it wasnโt the case.
From the logs:
Map on Screen:\n( 2, 0): ?\t( 3, 0): ?\t( 4, 0): ?\t( 5, 0): ?\t( 6, 0): ?\t( 7, 0): ?\t\n( 2, 1): ?\t( 3, 1): ?\t( 4, 1): ?\t( 5, 1): ?\t( 6, 1): ?\t( 7, 1): ?\t\n( 2, 2): ?\t( 3, 2): ?\t( 4, 2): ?\t( 5, 2): ?\t( 6, 2): ?\t( 7, 2): ?\t\n( 2, 3): ?\t( 3, 3): ?\t( 4, 3): ?\t( 5, 3): ?\t( 6, 3): ?\t( 7, 3): ?\t\n( 2, 4): ?\t( 3, 4): ?\t( 4, 4): ?\t( 5, 4): OBJ_1_1\t( 6, 4): ?\t( 7, 4): ?\t\n( 2, 5): ?\t( 3, 5): ?\t( 4, 5): ?\t( 5, 5): ?\t( 6, 5): ?\t( 7, 5): ?\t\n( 2, 6): ?\t( 3, 6): ?\t( 4, 6): ?\t( 5, 6): ?\t( 6, 6): ?\t( 7, 6): ?
Is anyone willing to demo a sample submission?
About 3 hours agoYou can skip all the steps above by just using the datasets from organizers: aicrowd/ChessExplained ยท Datasets at Hugging Face