WHAT NEW IN X11.01 SUN

AI collections from X · 22 collections · 64,204 posts

jevPOST

Jev + Jevcache + Fable benchmark: 220–780ms per decision

The author compares Jev's loop decision cost with Claude's, reporting decisions roughly 50x cheaper, Fable consulted only on three low-confidence steps, and a 26% faster repeat run with Jevcache.

Varun
@varun_mathurArchived 2026.09.21Originally posted 2026.09.21
Varun @varun_mathurOriginally posted 2026.09.21
Open original on X

noticed below cumulative impact of using Jev + Jevcache + Fable: - Per-decision: ~220–780ms Jev vs 20–40s claude before - Loop's decisions are ~50× cheaper; Fable was consulted only on the 3 low-confidence steps - jevcache: earlier identical-phrasing re-run showed cached (0ms replay) live.. repeat run was 26% faster (7.2→5.3 min)

This post quotes another post: x.com/varun_mathur/status/2101371145521905688

x.com/varun_mathur/status/2101740594754093349Archived 2026.09.21 10:14