Indie chatbot Auro V8 adds cross-session memory, claims self-iterating training loop
TheMoonMidas · x · 2026-10-06
Indie developer Alan (@aisentience) rolled out updates to Auro, a personality-focused chatbot: a post-training breakthrough produced the V8 model with higher scores on internal "Aura" and Emotional Intelligence benchmarks, and a new harness adds cross-session memory so the model learns about you over time. He also says Auro is now used to train its own next version, and blind tests already rank its "aura" above other models. Each message can return two hidden versions for blind voting, with an arena and API access. All metrics are self-reported; it's a personal project, not affiliated with OpenAI or Google.
More from Models
- Reflection AI debuts Beam: open agentic model with 501B total, 23B active params — bigblueboo · 2026-10-06
- Gemini 4 Argon tops Text Arena, writes a novel with 1M-token output limit — DynamicWebPaige · 2026-10-06
- Vision models beat Wordle from screenshots alone: OCR plus reasoning lands correct guesses — maximelabonne · 2026-10-06
- Google's Arohan: Gemini Nano chapter added to report, real inflection lands next year — _arohan_ · 2026-10-06
- LenVM lands COLM 2026 Spotlight: value model predicts remaining generation length for efficient reasoning — xwang_lk · 2026-10-06
- One model runs multiple games in parallel, researcher calls it bigger than LLM text generation — maximelabonne · 2026-10-06