Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).
Mainstream LLM chat tools converge on a linear conversation model, but users routinely consult several models on the same question, switch between specialists within a single task, run parallel analytical branches for comparison, and return to earlier turns to explore alternatives — work that is tree-shaped, not list-shaped. We present AIbranch, an open-source platform for branched, multi-model LLM conversations. Two interaction primitives are paired: a forkable conversation tree in which earlier turns remain navigable, and in-thread switching between a wide range of models — from commercial frontier models to locally-hosted ones — with cross-model context carryover. The architecture was designed for reusability and modularity, with a hub-plus-engines decomposition. The code base is released under the Apache License 2.0.
No comments yet — start the discussion below.