First Jev waitlist invitations go out
TypeSafe says it emailed a first batch of waitlisted users and will invite more people as access expands.
Jev latency and accuracy compared with Llama and DeepSeek
A user shares early results from trying Jev: with the same labeled inputs, Jev shows 235ms median latency and 82% accuracy, versus Llama 3.1 8b at 497ms/68% and DeepSeek v4 flash at about 2s.
(1/5) trying out jev this week with all the hype. kept it apples to apples vs the fast models it competes with: llama 3.1 8b and deepseek v4 flash same labelled inputs. jev: 235ms median, 82% accurate. llama: 497ms, 68%. deepseek flash was accurate but ~2s genuinely cool. then
· 3 likesOpen on X
More like this
User gains Jev access and runs ~5,000 requests
The user says they got Jev access earlier that day, ran ~5,000 requests costing about $2 across classification, model routing, intent, steering and more, and calls Jev a new intelligent decision-making primitive.
Jev Waitlist Open: Reply to Get Access
Kevin Zhang invites Jev waitlist users to reply with what they want to build to get access.