Skip to content

Add Legend of Elya — N64 dungeon crawler with AI NPCs#116

Open
Scottcjn wants to merge 1 commit intobobeff:mainfrom
Scottcjn:add-legend-of-elya
Open

Add Legend of Elya — N64 dungeon crawler with AI NPCs#116
Scottcjn wants to merge 1 commit intobobeff:mainfrom
Scottcjn:add-legend-of-elya

Conversation

@Scottcjn
Copy link
Copy Markdown

Summary

Adds Legend of Elya to the Role-Playing games section.

Legend of Elya is the world's first LLM running on Nintendo 64 hardware — an 819K-parameter nano-GPT transformer doing live inference on the MIPS R4300i CPU at 60 tokens/second. It's a Zelda-style dungeon crawler where NPCs respond dynamically using on-device neural network inference.

  • Written in C using the libdragon SDK
  • Full source code available
  • Produces a playable .z64 ROM
  • Uses the N64's RSP vector unit for matrix multiplication

Comment thread README.md

- **[Veloren](https://www.veloren.net/)** - multiplayer voxel RPG written in Rust. [[source]](https://github.com/veloren/veloren)

- **[Legend of Elya](https://github.com/Scottcjn/legend-of-elya-n64)** - World's first LLM on Nintendo 64. 819K-parameter nano-GPT transformer running live inference on the MIPS R4300i at 60 tok/s. A Zelda-style dungeon crawler with AI NPCs on 1996 console hardware. [[source]](https://github.com/Scottcjn/legend-of-elya-n64) **SDK: [libdragon](https://github.com/DragonMinded/libdragon)**
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The description for this entry is a bit longer and more promotional compared to others in the list.

To maintain consistency across the README, consider shortening it and using a more neutral tone similar to other entries.

For example:
"Zelda-style dungeon crawler for Nintendo 64 featuring AI NPCs powered by a nano-GPT model."

This would improve readability and keep formatting consistent.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants