Google Launches Gemini 4 Neon for Enterprise Work
Google DeepMind released Gemini 4 Neon, a new AI model positioned for enterprise, analytical, and scientific work. The model scores 61 on the Artificial Analysis Intelligence Index at its high thinking level and ranks second on the Vals Index. Neon is priced at $3 per million input tokens and $15 per million output tokens, with a 1 million-token context window, and is available today across Google's API, AI Studio, and consumer apps.
TL;DR
- Gemini 4 Neon ranks #3 overall on Artificial Analysis Intelligence Index (score 61) and #2 on Vals Index (70.9%)
- Model achieves top performance on enterprise benchmarks including Legal Research Bench, Tax Agent Bench, and Zapier's AutomationBench (56.3%)
- Pricing set at $3 per million input tokens and $15 per million output tokens, with 95% discount on cached input tokens ($0.15 per million)
- Ranks #2 on steerability for mid-task user corrections and records lowest fabrication rate in internal testing
Why It Matters
Gemini 4 Neon represents a competitive positioning move in the frontier AI market, targeting knowledge workers and enterprises with a model that balances capability and speed. The emphasis on steerability and low fabrication rates addresses practical deployment concerns beyond raw benchmark performance. Pricing and availability across multiple platforms signal Google's strategy to embed AI across consumer and enterprise workflows.
Business Impact
For enterprises, Neon's performance on domain-specific benchmarks like Legal Research Bench, Tax Agent Bench, and financial modeling tasks suggests direct applicability to professional services and back-office automation. The 1 million-token context window and multi-modal capabilities enable processing of complex documents and data in a single pass, reducing workflow fragmentation. Competitive pricing on cached tokens (95% discount) lowers costs for repeated queries on static documents.
Key Implications
- Google is competing directly on speed and cost efficiency, not just raw capability, with cached token pricing designed to reduce operational expenses for high-volume enterprise use
- Strong performance on specialized benchmarks (legal, tax, finance, spreadsheet modeling) indicates targeted optimization for professional services and knowledge work verticals
- Emphasis on steerability and low fabrication rates suggests the market is moving beyond benchmark scores toward practical reliability metrics for production deployment
What to Watch
Monitor adoption rates across Google's API and enterprise channels to assess whether Neon gains traction in professional services and financial services sectors. Track how competitors respond to the pricing model, particularly the cached token discount structure. Watch for updates on fabrication rates and steerability performance as real-world usage accumulates.
Subscribe to the newsletter
The latest stories and analysis, delivered to your inbox.
Free. No spam. Unsubscribe any time.