CUA Driver Bench: cua-driver hits 58% on macOS background desktop tasks vs Codex 33%
victor_explore · x · 2026-11-07
Cua Driver Bench (with Scale's AgentEnv) measures whether computer-use drivers can finish real desktop tasks while you keep working, without disturbance or shell cheats. On macOS full background passes: cua-driver 58% (23/40), Codex Computer Use 33%, vercel agent-desktop 20%, browser-use 3%. On Windows cua-driver leads at 28%. Key finding: much of the gap is disturbance, not failure; background dialogs work on macOS but not yet Windows. Tasks, runner and scorer are public for reproducibility.
Related event: Cua Launches Driver Bench for Background Desktop Agents(2 posts)→
More from coding & agent
- libghostty 1.0 ships, already powering terminals for 12.5M JetBrains users — jedisct1 · 2026-11-07
- Teknium: Context retrieval beats alternatives once sessions exceed 100K tokens — Teknium · 2026-11-07
- Claude Opus 5.5 turns 3-minute task into 3-hour fiasco with 57 self-logged mistakes — PMinervini · 2026-11-07
- Xepak: declarative local builder for REST, RPC, MCP and RAG servers that agents can write DSL for — rumatoest · 2026-11-07
- Dev ports T3 Code to the Nintendo 3DS on a whim — Josikinz · 2026-11-07
- Dev distills Anthropic's Renaissance engraving art style into a reusable image-generation Skill — yihui_indie · 2026-11-07