Gemini 4 Neon rolls out across Google AI products
Google announced Gemini 4 Neon on November 18 with a one-million-token context window. Listed API prices are USD 3 per million input tokens, USD 0.15 for cached input and USD 15 for output. Rollout includes AI Studio, Gemini for business, and Pro or Ultra access in Antigravity and the Gemini app. Access remains subject to product eligibility and limits. GoToAI has not independently tested performance.
RTX Spark compact desktops launch from three manufacturers
NVIDIA announced availability from ASUS, Dell and MSI on November 18, with Acer, GIGABYTE, HP and Microsoft systems to follow. Configurations offer up to 128 GB unified memory. Supported local processing can avoid per-token cloud API charges for those tasks; this does not remove purchase, power or software costs. The announcement does not establish local stock or a universal price. GoToAI has not tested the machines.
OpenAI explains its internal Performance Factory
OpenAI published the case study on November 19. Its internal system assigns Codex agents to investigate performance alerts and propose changes, with engineers reviewing and deploying fixes. Agents subsequently verify recovery. OpenAI reports that some root-cause suggestions need refinement, and fixing regressions remains harder than diagnosing them. This is an internal engineering account, not a public product launch or independently audited reliability result.
How this edition was made
AI-generated narration with original editorial illustrations and English captions. Sources checked 2026-11-20. Technical audio checks and transcript review were completed; subjective pronunciation and naturalness were not independently assessed.
Check the sources
AI-assisted reporting, checked against the sources linked above. This article describes announcements unless a hands-on test is explicitly identified. Our editorial standards.
