Cost-vs-quality chart puts Gemini 4 Neon on the CursorBench Pareto frontier
zacharynado · x · 2026-11-20
Quoting a tweet complaining that Gemini 4 Carbon still can't be tested and Argon was disappointing, the author shares a CursorBench 4.1 score vs average cost-per-task chart: Gemini 4 Neon sits at the frontier with 65%, ahead of Carbon/Hydrogen, Opus 5.5, Sonnet 5.5 and Grok 4.7, while GPT-5.6 Sol and Muse Spark 1.3 trail at the low-cost end. Takeaway: skip Carbon, just use Neon.
More from coding & agent
- Vercel's new UI SDK draws praise from shadcn for building draggable calendar, date picker — shadcn · 2026-11-20
- Dev shows AI meal-planning tool that matches macros using whole grocery units — lucasmeijer · 2026-11-20
- Agents run animal content accounts to 300M-impression marketing playbook — Saul_Loveman · 2026-11-20
- Dev ships agent-built recipe site with copyable agent instructions — lucasmeijer · 2026-11-20
- Codex bug: GPT-6.2 cache hit rate drops from 97.4% to 46.4%, system prompts billed at full price — steipete · 2026-11-20
- 40 lines of zero-dependency Node: AES-256-GCM encrypt env vars before your PaaS leaks them — DavidWells · 2026-11-20