Google's Ammaar confirms Gemini 4 Carbon hits the Pareto frontier on Cursorbench
ammaar · x · 2026-11-11
Ammaar, who works at Google, replied confirming the Gemini 4 Carbon claim: the model sits on the Pareto frontier of Cursorbench (performance-vs-cost). The quoted tweet reads "Gemini 4 Carbon at the Pareto on Cursorbench! Hope you all enjoy building with it" — phrasing close to an official launch note. This is an inside confirmation of the next-gen model's coding performance; detailed scores are not yet public.
Related event: Gemini 4 Carbon Tops CursorBench 4.0 Upon Cursor Launch(5 posts)→
More from Models
- Training a 'is this joke funny' model is hard; targeted critique prompts work better — ivan_bezdomny · 2026-11-11
- 1.3B open-source MiniCPM-OCR 1.0 tops Real5-OmniDocBench, beating GPT-5.2 — NielsRogge · 2026-11-11
- Claude writes dark, offensive jokes freely while GPT censors them, per hands-on test — ivan_bezdomny · 2026-11-11
- Dev trains a 944K-param autocomplete model on Enron emails for 35 cents — victormustar · 2026-11-11
- Google Cloud free trial now excludes Gemini platform, including Vertex AI and API access — durzagon · 2026-11-11
- Did SuperGrok Cut Limits Again? 900k Tokens Burned on Cold Starts and Cache Misses — NickPassig · 2026-11-11