{"slug": "google-releases-gemini-4-neon-at-3-15-scoring-61-on-aa-index", "title": "Google releases Gemini 4 Neon at $3/$15, scoring 61 on AA index", "summary": "Google released Gemini 4 Neon at $3 per million input tokens and $15 per million output tokens, scoring 61 on the Artificial Analysis Coding Agent Index, according to Artificial Analysis. The index score puts Gemini 4 Neon in line with Claude Code on Fable 5.2 at roughly half the cost, and its $6.44 per task is about 17% below Gemini 4 Carbon. Coding agent teams get comparable results for about half the spend.", "body_md": "Artificial Analysis puts its Coding Agent Index score in line with Claude Code on Fable 5.2 at roughly half the cost, and its $6.44 per task is about 17% below Gemini 4 Carbon. Coding agent teams get comparable results for about half the spend.\nRead: Artificial Analysis puts its Coding Agent Index score in line with Claude Code on Fable 5.2 at roughly half the cost, and its $6.44 per task is about 17% below Gemini 4 Carbon. Coding agent teams get comparable results for about half the spend.\nRead: Agent jobs that can wait, such as overnight reports, now get up to 40% off GPT-6.2 Astra and Sol inside the Responses API, with no separate batch pipeline to build. Reusing a continuation ID also moves follow-up requests up the queue.\nTry: A week after six fixes to the same servers, mcp-server-git still let a repository's own config run commands with the server's privileges during a merge or read. Agents that open untrusted repositories through it carried that risk until this release.\nRead: Slower inference explains only part of why Carbon runs take longer than Opus and Astra runs. Most of the gap is extra tool calls and re-sent context, about 16x the input tokens, so Carbon's per-token price understates what each fixed bug costs.\nRead: Human review found real defects in about half of the flagged patches, so a passing test suite overstates how often an agent's fix is correct. BenchProofer builds formal specs and Lean proofs from existing tests, a stricter check teams can run on agent patches.\nRead: Human auditors scored AUROC 0.59, near chance, and the best frontier model reached 0.73 and fully identified 32% of cases. Handing research to agents and reviewing their transcripts afterward rests on a check that misses most sabotage.\nRead: Microsoft AI released seven models across code, cyber, image, voice and reasoning. Agents pay per token at every step, so a coding model at 40% of Sol's price cuts that bill by more than half, if outside tests confirm Microsoft's own quality figures.", "url": "https://wpnews.pro/news/google-releases-gemini-4-neon-at-3-15-scoring-61-on-aa-index", "canonical_source": "https://www.vibeleaderboard.ai/intel/brief/2026-11-19", "published_at": "2026-11-19 12:08:29+00:00", "updated_at": "2026-11-19 12:42:11.647377+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-agents", "ai-products", "ai-tools"], "entities": ["Google", "Gemini 4 Neon", "Artificial Analysis", "Claude Code", "Fable 5.2", "Gemini 4 Carbon"], "also_reported_by": [], "takeaway": "Gemini 4 Neon scores 61 on the Artificial Analysis Coding Agent Index at $3/$15 per million tokens and $6.44 per task, about 17% below Gemini 4 Carbon, matching Claude Code on Fable 5.2 at roughly half the cost.", "recommendation": {"advice": "Route coding-agent work to Gemini 4 Neon where its 61 index score matches Claude Code on Fable 5.2, since it costs roughly half as much and $6.44 per task, 17% below Gemini 4 Carbon.", "playbook": "llm-cost", "playbook_url": "https://wpnews.pro/playbooks/llm-cost"}, "alternates": {"html": "https://wpnews.pro/news/google-releases-gemini-4-neon-at-3-15-scoring-61-on-aa-index", "markdown": "https://wpnews.pro/news/google-releases-gemini-4-neon-at-3-15-scoring-61-on-aa-index.md", "text": "https://wpnews.pro/news/google-releases-gemini-4-neon-at-3-15-scoring-61-on-aa-index.txt", "jsonld": "https://wpnews.pro/news/google-releases-gemini-4-neon-at-3-15-scoring-61-on-aa-index.jsonld"}}