GPT-6.2 Sol tops browser-use leaderboard at 82%, 35% cheaper and 2x faster than predecessor
charliermarsh · x · 2026-11-18
browser-use published its latest browser-agent leaderboard: GPT-6.2 Sol ranks #1 with an 82% success rate, up from 81% for GPT-6.1 Sol.
Key numbers:
- 35% cheaper than the previous version
- 2x faster: 12 minutes instead of 22
- 36% fewer steps
- charliermarsh amplified it: OpenAI is "by far SOTA for browser use," 2x cheaper than Opus 5.5 and 7 points better
More from Models
- Decision models beat LLMs at structured output: Liquid AI d1 runs 2.7-3.6x faster than LFM3 — mervenoyann · 2026-11-18
- Why Anthropic matters: competition, pricing and open source — dh7net · 2026-11-18
- Dev's Deep Dive: GPT-6.2 Astra Improved but Still 'Directionally Very Bad' at Debugging — kimmonismus · 2026-11-18
- OpenAI Charged for a ChatGPT Business Premium Seat for a Month That Can't Be Used — physmathastro · 2026-11-18
- Hunch: Astra 6.2's bad reviews stem from OpenAI dogfooding with unlimited internal swarms — daniel_mac8 · 2026-11-18
- Google rolls out Antigravity Orchestrator: dynamic in-task model switching, 24% less quota — rseroter · 2026-11-18