Show HN: Work at a Fortune 500 and never run into token limits? Now you can

A developer has created a tool called Vibe that allows users to work with large language models without hitting token limits, even when using Fortune 500-level workloads. The tool supports multiple models with up to 200k context windows and is available as an open-source project.

vibe s ~/projects/vibecode hacker v0.7.3 · build 41a9c2 vibes v0.7.3 · node 20.12 · darwin-arm64 ⊘ viewing a historical session — replies are frozen and the input is disabled ↪ go to current › ⏎ send ⇧⏎ newline ⌃C interrupt ⇥ complete ↑↓ history / commands ⌘K sessions vibe-7 · prism ● booting │ ~/projects/vibecode │ main │ ctx: 0 / 200k tokens: — │ utf-8 · LF │ — select a model read-only vibe-7 · prism balanced · 200k ctx vibe-7 · flux max reasoning · 200k ctx vibe-7 · swift fastest · 200k ctx vibe-5 · classic deprecated · 100k ctx esc to close selection takes effect on next request jump to session ⌘K esc to close