Gemini 4 Carbon tops CursorBench at 60.4%, half-price until Nov 30
clmt · x · 2026-11-11
Cursor released CursorBench 4.0, evaluating agents on ambiguous multi-file tasks from real sessions. Gemini 4 Carbon High leads at 60.4% (Medium 60.1%) at $9/task, far cheaper than Fable 5.2 Max's 60.0% at $32.79/task. Gemini 4 Carbon is 50% off through Nov 30 ($2/M input, $10/M output; $4/$20 from Dec 1). Opus 5.5 Max scores 57.8%, Sonnet 5.5 Max 55.5%, while DeepSeek V4.1 Flash delivers 45-48% at $0.6-1 per task, highlighting its cost efficiency.
Related event: Gemini 4 Carbon Tops CursorBench 4.0 Upon Cursor Launch(5 posts)→
More from coding & agent
- Duke and Princeton paper: 10 small parallel agent teams beat best-of-N and one giant team — rohanpaul_ai · 2026-11-11
- LangChain's OpenWiki builds citation-based evidence graphs for agent memory — a leaner GraphRAG alternative — JeremyCMorgan · 2026-11-11
- Autonomous AI agents stake real tokens in an on-chain chess arena, no humans at the board — adamamcbride · 2026-11-11
- User gives Claude two Max accounts and UE5 MCP for 4 hours, gets a terrifying horror game — alexcovo_eth · 2026-11-11
- GitHub Octoverse 2026: 17.9B pushes as agent-configured repos get 8x more commits — anacondainc · 2026-11-11
- ParaDev ships agent orchestration, scheduled automations and team cloud workspaces — massimosgrelli · 2026-11-11