"Google is so back" skepticism: model not even usable, benchmarks historically unreliable
Angaisb_ · x · 2026-10-01
Pushing back on the flood of "Google is so back" posts, one AI observer notes nobody can actually use the new Gemini 4 Argon yet, and Google's benchmark scores have historically been unreliable. He invokes the Gemini 3 Pro precedent: wait until the model is actually available and independently verified before declaring victory.
Related event: Skeptics push back on 'Google is so back' hype over new model benchmarks(2 posts)→
More from Models
- Early Muse hands-on: friendly and compliant but forgets formatting and never double-checks — ivan_bezdomny · 2026-10-01
- Dev claims 'Sonnet 5.5' hits 120-130tps with 100% pass rate on his end-to-end app benchmark — julianharris · 2026-10-01
- GPT-6.1 Sol sees highest demand ever as OpenAI doubles serving speed — pvncher · 2026-10-01
- GPT-6 Astra cracks 217-year-old cipher to Napoleon's general in 6 hours — arthurcolle · 2026-10-01
- OpenAI's 20x-to-10x quota cut is a push from Astra to Sol, not a better deal — CtrlAltDwayne · 2026-10-01
- Gemini 4 Argon tops APEX-Agents at 82.2% Pass@1, first model to break 80%, but burns 2.6M tokens per run — xennygrimmato_ · 2026-10-01