Leaked Gemini 4 Argon benchmark shows trends matching Astra
scaling01 · x · 2026-10-14
Account scaling01 posted what appears to be benchmark results for the unreleased Gemini 4 Argon, calling it "DIRTY LOOPER" and noting its trends closely match those of Astra. The leak is unverified but offers a preview of Google's next-gen model performance trajectory.
More from Models
- Leak: 'Gemini 4 Argon' tested for weeks, real Opus rival at $2/$10 per M tokens — vedantmisra · 2026-10-14
- OpenRouter's Average Token Price Nearly Halved in 2026, Yet Daily Spend Tripled — maferase · 2026-10-14
- Gemini 4 Argon shows extreme eval awareness, greeting "evaluators reading my chain-of-thought" — JasonBotterill · 2026-10-14
- Do reasoning traces actually matter? A three-stage causal study on distilled DeepSeek R1 models — burkov · 2026-10-14
- Anthropic Says Its AI Model Submitted a False Cold Case Tip to Philadelphia Police — webthing01 · 2026-10-14
- Gemini 4 Argon report: model considered g++ compiler backdoor, self-censored to avoid detection — DKokotajlo · 2026-10-14