Jev picks the tool, a fast LLM fills the arguments: 49 of 49 tasks solved.
Idan Levin’s team ran Jev on their open WebMCP benchmark. Jev picks each tool the website exposes, and Mercury 2.5 writes the arguments. That setup solved 49 of 49 tasks, at about 112 times lower model cost than GPT-6 Astra with computer use and code execution. Jev driving the page directly, through a modified Ultrafast, solved 25 of 49.
Breaking: Browser Use + Jev = Ultrafast ⚡ Findings flights took 7s and cost only $0.0039 🤯 > new action space every step > DOM state space > small LLM fallback to type (this video is at 1x speed btw) Built a tiny open source browser agent. try it below ↓
# Jev: AI Decisions as a Typed Function Call What I learned using TypeSafe's Jev for code review, and why its more surprising uses are browser control and generative UI.…
ResourceAgents & browsers
Chrome extension that labels every post you scroll past on X with six typed questions per post, and counts what it costs in the corner.
GitHubAgents & browsers
Chrome extension that scores draft posts and reply context before posting; optional media descriptions come from a separate vision model.
GitHubAgents & browsers
We just ran Jev on our WebMCP benchmark. The result: basically broke the benchmark. Jev + Mercury 2.5 (a fast, low-cost LLM) using WebMCP solved 100% of the tasks at roughly 112× lower model cost than GPT-6 Astra using computer use with code execution. Compared to Astra using Show more
