Hands-on with GPT-6.2: Ignore the hate, it's a cheap computer-use workhorse
nickbaumann_ · x · 2026-11-18
Alex Finn pushes back against the wave of GPT-6.2 hate on X, arguing that X discourse isn't real life.
His own testing:
- Demos and benchmarks he ran were genuinely impressive — better than Astra 6 at lower cost;
- Results fall short of Fable 5.2, and it's not close to the leap Astra 6 was;
- Still, it's an efficient super intelligence that is incredible at computer use with low running costs.
His advice: don't let the open-letter crowd set your opinion — test it yourself, build your own benchmarks, run it against your existing workflows, and you'll find real value.
Related event: Mixed Reactions to GPT-6.2 as Hands-On Tests Push Back on Criticism(2 posts)→
More from Models
- GPT-6.2 Astra not obviously better than GPT-6 at Blood on the Clocktower eval — SamuelAlbanie · 2026-11-18
- Gemini meeting transcripts auto-mask swearwords, HF researcher plans to stress test it — _lewtun · 2026-11-18
- GPT 6.2 Astra and Sol Take 4th and 5th on sudo L7 Coding Benchmark, Still Behind Latest Claude — echen · 2026-11-18
- Claude leads Arena agent leaderboard with nearly 2x praise-to-complaint ratio — thursdai_pod · 2026-11-18
- GPT-6 Cyber uncovers critical Java bug affecting ~90% of Fortune 500, Calif says — gdb · 2026-11-18
- GPT-6.2 Astra panned by top reviewers: 'essentially unusable,' falls short of Opus 5.5 and Fable — deedydas · 2026-11-18