Cloud agentsLet AI research in the cloud — handling multi-step tasks or running on a schedule.
Research skillsLet AI pick the right methods and tools for each task, and extend it through the skill market.
Paper setsSearch, filter and combine papers, then analyze within a clearly scoped body of material.
Research panelSee sources, papers and results beside the conversation — check the evidence and steer as you go.

Literature discoveryStart from a research question, find the relevant papers, and build a list worth reading closely.
Topic reviewsMap technical approaches, compare evidence, and turn many papers into a topic report.
Ongoing trackingFollow a direction over time, with sourced research briefs and papers queued to read.
Topic researchBring topic materials and results together so AI can watch progress, revisit what you know, and suggest next leads.

Benchmarks

1 report

DigestsOctober 14, 2026

Multimodal Data & Benchmarks: August 2026 Progress Brief

August 2026's multimodal benchmark literature is unusually convergent. Rather than announcing new tasks, it rebuilds evaluation around the specific step where current models break: an unanswerable clip a model will not flag, a second image it cannot keep straight, a polygon it…

Multi-modal Benchmarks

Benchmarks - Parallax