Unverified Gemini 4 Carbon eval sheet claims wins on 14 of 18 benchmarks vs Claude and GPT
Angaisb_ · x · 2026-10-29
An unverified eval sheet linking to deepmind.google circulates online: Gemini 4 Carbon reportedly scores 59 on AA Index v5.0, edging Claude Opus 5.5 (58), and leads on 14 of 18 benchmarks including SWE-sweep (9.6%), Vending-Bench 2 ($16,484), GraphWalks (96.2%) and SEC-bench Pro (88.2%), against Gemini 4 Argon, GPT-6 Astra, Claude Opus 5.5 and Fable 5.1. GPT-6 Astra tops FrontierSWE 2 at 65.5%. Treat with skepticism — no official confirmation.
Related event: Google Launches Gemini 4 Carbon, Topping AA Index Two Weeks After Argon(82 posts)→
More from Models
- Google Announces Gemini 4 Carbon, Its Next Frontier Model, Starting With Cyber Defenders — Snoo26837 · 2026-10-29
- Suhail launches Particle: particle-1 matches performance with ~45% fewer tokens — ycombinator · 2026-10-29
- Defenders of Google's pace point to Genie 4 world model launch and Gemini 4 Argon announcement — haider1 · 2026-10-29
- 'I didn't realize Argon was the mid-sized model': DeepMind naming surprises users — saurabh_shah2 · 2026-10-29
- Unverified 'Gemini 4 Carbon' output showcase thread circulates ahead of any release — cgarciae88 · 2026-10-29
- Rant on confusing model codenames, plus claim that "barium" was shelved for being too spiky — unironictechbro · 2026-10-29