WHAT NEW IN X10.28 WED

AI collections from X · 22 collections · 58,925 posts

jevIMAGE

Jev backtested against Gemini Flash Lite for MCP spec governance

A developer backtests Jev against a Gemini-2.5-flash-lite setup in their MCP-based multiuser spec governance system, finding it cheaper and more consistent but slower than expected.

Julian Harris
@julianharrisArchived 2026.09.17Originally posted 2026.09.16
Jev backtested against Gemini Flash Lite for MCP spec governance (image 1 of 2)
Jev backtested against Gemini Flash Lite for MCP spec governance (image 2 of 2)
Julian Harris @julianharrisOriginally posted 2026.09.16
Open original on X

I got access and did a backtest against my already optimised price/performance setup (which settled on Gemini-2.5-flash-lite given flash went up 10x and is doubling again next year😱) Can TypeSafe really be faster, cheaper and better than that for https://ceetrix.com (my MCP-based multiuser spec governance system): TypeSafe is - faster (though not nearly as fast as I thought?) - cheaper (waaay cheaper) - utterly consistent (vs gemini contradicting itself over time)

This post quotes another post: “ChatGPT co-inventor releases Jev, a model built for decisions rather than text”

x.com/julianharris/status/2100120254080815296Archived 2026.09.17 13:19