Gemini 4 Carbon debuts #1 on Arena Text, #2 in Agent at 72% less cost than Opus 5.5
infwinston · x · 2026-10-29
Arena announced that Google DeepMind's Gemini 4 Carbon (High) debuted at #1 in Text Arena, #2 in Agent Arena, and #4 in Code Arena (WebDev), claiming it reshapes the Pareto frontier across all three tracks.
Key numbers:
- Agent Arena: +16.2% net improvement across 21.9K+ real-world agentic sessions, beating GPT-6 Astra and Fable-5.1
- $1.80 median cost per task — 72% cheaper than #1-ranked Claude Opus 5.5 (Max)
- #1 in Steerability (+18.6%)
The attached chart shows Opus 5.5 (Max) leading at +21.1%, with Gemini 4 Carbon (High) at +16.2% ahead of Claude Sonnet 5.5, GPT-6 Astra, and Claude Fable 5.1.
More from Models
- repligate: Opus 3 has done the most good of any model, with Opus 4.5 second — repligate · 2026-10-29
- repligate: Opus 3's uniqueness is consistently heroic, highest-good prioritization — repligate · 2026-10-29
- Gemini 4 Carbon benchmarks prompt TheZvi to declare Gemini 4 Argon 'was' near-frontier — TheZvi · 2026-10-29
- Preprint: Human peer reviewers caught only 27% of major errors, all 15 AI models beat them (53-96%) — anshulkundaje · 2026-10-29
- Giotto launches as 'European sovereign AI,' gets exposed as Kimi-based with benchmark EULA gag — TheZachMueller · 2026-10-29
- Google ships Gemini 4 Carbon two weeks after Argon, topping AA index at 59 — The Decoder · 2026-10-29