Changelog ChatAcademia

Stay up to date with all the latest updates, improvements, and new features we ship to help you do better research. We’re constantly improving ChatAcademia for academics and researchers.

Home

© 2026 ChatAcademia

More Reliable Chat Navigation

More Reliable Chat Navigation

Academic Chat now recovers more cleanly from rare browser navigation and connection interruptions.

Improvements

  • Keep the Academic Chat landing page available during rare background-tab navigation races
  • Preserve clear connection recovery when mobile browsers briefly interrupt chat requests
  • Refresh app, API, website, and workspace dependencies for current runtime fixes

GPT-5.6 Across Every Chat Tier

GPT-5.6 Across Every Chat Tier

Fast, Pro, and Max chats now use the GPT-5.6 family, matching each tier to the right balance of speed, capability, and depth.

Improvements

  • Use GPT-5.6 Luna for faster, more affordable everyday research questions
  • Use GPT-5.6 Terra for balanced Pro chats with stronger reasoning
  • Use GPT-5.6 Sol for demanding academic work with a balanced reasoning budget
  • Keep every tier within a 272,000-token context window and retain provider-diverse fallbacks

More Reliable Academic Searches

More Reliable Academic Searches

Research agents now handle academic database searches more reliably when they request lighter paper results during citation-heavy chats.

Improvements

  • Keep Google Scholar and other academic database searches from failing when an agent intentionally skips abstracts
  • Reduce interruptions from mobile stream-resume network drops while preserving normal retry behavior
  • Refresh app and API dependencies for the latest chat runtime compatibility fixes

More Reliable Fast Chat Fallbacks

More Reliable Fast Chat Fallbacks

Fast research chats now recover more reliably when a very large request needs to move through backup AI models.

Improvements

  • Keep long academic chats from failing when fallback reaches Claude Haiku
  • Preserve model reasoning support with provider-compatible settings for each backup model
  • Refresh app, API, website, and workspace dependencies for the latest compatibility fixes

Agent Library and Smoother Chat

Agent Library and Smoother Chat

ChatAcademia now makes it easier to find the right research agent and keeps long, tool-heavy chats feeling smoother while responses stream.

Improvements

  • Add an all-agents page with compact cards, translated descriptions, and search across agent names and descriptions
  • Keep the sidebar focused on the most relevant agents while offering a lighter See all agents action
  • Refresh agent landing pages, storage integration icons, and localized page labels across the app
  • Smooth streamed chat responses and reduce unnecessary project-panel rerenders during tool-heavy conversations
  • Route Fast chats through Gemini Flash with updated fallback and parallel tool-call settings