TheirSide resource desk
Evaluation records and field tools
Check the memory and group-chat records, or plan a fictional character with a no-login template.
Resource index reviewed
Evaluation v0.1 records
Two fixed demos, an isolated Chinese memory test with 21 checks, and one controlled 10-run production record, with methods, failure criteria, observed results, and explicit limits.
- 01
AI character memory: a cross-session test and fixed demo
A September 5, 2026 synthetic individual character chat test passed 21 of 21 checks across three trials for immediate Chinese recall, correction and negative cases. The earlier fixed demo is retained separately; neither record establishes long-term retention or Production performance.
- 02
Can one AI character respond to another? A 10-run production test
A dated production test ran one fixed English prompt in 10 fresh signed-out guest rooms with Mio and Kai. All 10 flows completed with Mio followed by Kai, and all 10 later Kai replies explicitly named Mio and continued from a run-specific idea before adding a complication.
- 03
What a scripted AI story scene establishes at the opening
A fixed bilingual story-page demo documents the visible opening state of a scripted scene—status, role context, known facts, and two character lines—without predicting later branches or endings.
Templates
Practical browser-local tools you can use without sending draft content to TheirSide.
- 01
AI character profile template: a no-login field workbook
Shape a fictional character across nine practical fields, edit a complete example in browser memory, then copy versioned Markdown or download Unicode-safe JSON without creating an account.