Gemini 4 Neon tops Coding Agent Index
Google's Gemini 4 Neon has overtaken Anthropic's Claude Fable 5.2 on the Artificial Analysis Coding Agent Index, scoring 59 at extra-high reasoning effort. The mid-tier model delivers top-tier agentic coding benchmark performance at roughly half the operational cost and higher token throughput than rival frontier models.
Google's mid-tier model dethroning Anthropic's flagship on agentic coding proves that the efficiency-to-intelligence frontier is shifting rapidly. Developers no longer need to pay flagship token premiums to get state-of-the-art coding agent execution.
- –Leaderboard shift: Running Gemini 4 Neon at its highest reasoning setting (xhigh) achieved a score of 59 on the Artificial Analysis Coding Agent Index, edging past Claude Fable 5.2 (~58).
- –Cost efficiency: At $3/M input and $15/M output tokens, Neon provides top-of-the-leaderboard agent capability at roughly half the execution cost of heavy Claude configurations.
- –Throughput advantage: Clocking around 110 output tokens per second, Neon accelerates agent loops and noticeably reduces multi-turn debugging latency.
- –Immediate integration: With support across Google AI Studio, Google Antigravity, and third-party harnesses like Cursor, developers can deploy Neon directly into existing coding agent stacks.
DISCOVERED
6h ago
2026-11-19
PUBLISHED
6h ago
2026-11-19
RELEVANCE
AUTHOR
0x0SojalSec