Show HN: A Stateful LLM API that works across providers
TL;DR
For the last year we've been building AI workspaces. With every new product, we found ourselves rebuilding the same context layer and infrastructure. We tried existing solutions, but we'd either have to lock into one provider or have to manage the storage infrastructure. What we wanted was an API that's stateful and isn't tied to a specific provider.
Nauti's Take
Twigg is worth a focused prototype test for a small team: connect one existing workflow to two model providers and inspect tool calls, long histories, failure handling, and provider switching. Before production use, verify data residency, deletion controls, exportability, and the cost of storing each session, since those details are missing from the available summary.
Sources