AI Trends

Gemini 4 Neon (High) Ranks No. 2 Overall on the Agent Arena Index

ArenaUpdated Nov 19 09:28

AI summary

Gemini 4 Neon (High) ranks No. 2 overall on the Agent Arena Index with 1,336 pts, making it the highest-ranked Google DeepMind model. The result is based on 4.2k real-world agent sessions from the global Arena community, with a median cost of $2.82 per task. It scored higher than the No. 3 model, Claude Opus 5.5 (Max), while costing 53% less per task. It ranks No. 2, No. 2, and No. 3 in Praise vs. Complaint, Steerability, and Task Completion, respectively.

Why read this

SelectedAI score 76

Results from 4.2k real-world agent sessions, compared against per-task costs, provide a concrete reference point for assessing the balance between capabilities and costs for Gemini and Claude.

Topics

Tags

Quoted post

Google DeepMind

@GoogleDeepMind

Gemini 4 Neon is rolling out now. Sitting between Carbon and Hydrogen, it brings frontier intelligence and fast task completion for complex software engineering, knowledge work, and science.

View original

Gemini 4 Neon is rolling out now. Sitting between Carbon and Hydrogen, it brings frontier intelligence and fast task completion for complex software engineering, knowledge work, and science.

View original post

Body

Big news: Gemini 4 Neon (High) is the best @GoogleDeepmind model on the Agent Arena Index, ranking #2 overall and reshaping the Pareto frontier!

With 1336 pts based on 4.2k real-world agentic sessions from Arena’s global community, it surpassed: - #4 Carbon (High): 1303 pts - #9 Hydrogen (High): 1271 pts - #11 Argon (High): 1229 pts

Neon is on the Agent Arena Pareto frontier at $2.82 median cost per task. It scores higher than #3 Claude Opus 5.5 (Max) for 53% less per task, and is 69% less per task than #1 Claude Fable 5.2 (Max). See Pareto in thread.

It's strong across key signals: - #2 Praise vs. Complaint: 1539 pts (user satisfaction with the agent’s outcome) - #2 Steerability: 1298 pts (how well an agent responds to user corrections) - #3 Task Completion: 1316 pts (explicit confirmation of task success from users)

See where it landed in the Code Arena: WebDev and Text Arena below in thread.

Congrats to the team at @GoogleDeepmind on another huge release!

Source

Arena