AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
Documentation for fusionkit (@fusionkit/cli): run a panel of models behind Codex, Claude Code, or Cursor, and let a judge synthesize the answer your agent runs.
Developer toolsGo · TypeScriptGitHub @velum-labsOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
The coordination layer for hybrid distributed AI compute.
Taken from the repository README
Loading the full README. The saved excerpt is shown below.
FusionKit fuses a panel of models - including open-weight models at a fraction of frontier prices - behind your unmodified coding agent. A single turn fans out to the panel, captures candidate trajectories, runs judge + synthesis, and returns one native Codex / Claude Code / Cursor response; the harness never knows fusion happened.Read the full README ↗The full README is unavailable right now. Read it at the original link. The saved excerpt is shown below.
FusionKit fuses a panel of models - including open-weight models at a fraction of frontier prices - behind your unmodified coding agent. A single turn fans out to the panel, captures candidate trajectories, runs judge + synthesis, and returns one native Codex / Claude Code / Cursor response; the harness never knows fusion happened.Read the full README ↗