VFF - The signal in the noise
Model ReleaseTrending

Google Launches Gemini 4 Neon for Enterprise Work

Read original
Share
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.

  • 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

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.

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.

  • 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

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.

OneUpAI
OneUp Your Business. Get More Done. OneUp Your Business. Get More Done. OneUp Your Business. Get More Done.
Learn More
Share

Subscribe to the newsletter

The latest stories and analysis, delivered to your inbox.

Free. No spam. Unsubscribe any time.

Related stories

Amazon, Google to control 30% of Anthropic after IPO
TrendingNews

Amazon, Google to control 30% of Anthropic after IPO

Amazon and Google will collectively control nearly 30% of Anthropic following its IPO, with Amazon holding just under 15% and Google holding nearly 14%. Amazon has secured an option to increase its stake significantly beyond its initial position. The dual major investments from two of the world's largest tech companies underscore the strategic importance of AI infrastructure and large language model development.

by Julie Bort· TechCrunch AI
Google DeepMind Strategy Chief Joins Anthropic
TrendingNews

Google DeepMind Strategy Chief Joins Anthropic

Jasjeet Sekhon, chief strategy officer at Google DeepMind, has departed to join Anthropic. Sekhon announced the move on X, stating he is working to ensure Claude practices and values epistemic rigor. Anthropic confirmed the hire. The departure represents a notable shift of a senior strategy executive from Google's AI division to a competing AI safety-focused company.

by Ryan Barwick· The Information
Google Maps adds AI trip planning with custom itineraries
TrendingModel Release

Google Maps adds AI trip planning with custom itineraries

Google is rolling out an AI-powered trip-planning feature within its Ask Maps chatbot that generates custom itineraries with day-by-day schedules, stop lists, routes, and numbered pins. The feature works for both single-day outings and multi-day trips, with examples including holiday shopping days and three-day road trips with coastal hikes and dining stops. The update integrates Gemini AI capabilities directly into the Maps app to automate travel planning.

by Emma Roth· The Verge AI
Google Rolls Out Gemini 4 Hydrogen for Complex Search Queries
TrendingModel Release

Google Rolls Out Gemini 4 Hydrogen for Complex Search Queries

Google has released Gemini 4 Hydrogen, a new AI model designed to handle complex, multi-part questions with improved accuracy. The model is now the default in Google Search's AI Mode and available across Google's AI products including the Gemini app, AI Studio, API, and business offerings. According to Google Search VP Robby Stein, the model is being used for the hardest questions in AI Overviews.

by Stevie Bonifield· The Verge AI