Skip to content
DigIntoAI/Model intelligence
Model intelligence feed

Model intelligence.

Follow the changes. Understand what comes next.

Editor’s pickGoogle
Google/Gemini 4 Neon

Google Launches Gemini 4 Neon in Antigravity With 1M Context, Four Thinking Levels, and Benchmarks vs. Claude Opus 5.5 and GPT-6.2 Sol

Announced on November 19, 2026 at 05:51 Beijing time, Google's Gemini 4 Neon features a 1M-token context window, four thinking levels, and planner-executor pairing with Gemini 4 Hydrogen in Google Antigravity, outperforming Claude Opus 5.5 and GPT-6.2 Sol across most of 18 published benchmarks.

Latest stories349

Newest published
97

Vercel Security Review: Reviews Free Until November 10, Fixes Billed Separately

Vercel’s November 3 announcement says Security Review is free until November 10, 2026, for its Pro and Enterprise public beta; Vercel Agent fixes remain separately billed. Owners enable the service, GitHub pushes trigger reviews that normally examine changes, and connected-repository fixes become pull requests for human review. Regular pricing adds $0.25 per million tokens to provider inference costs. The announcement gives no cutoff time or time zone.

Vercel Security ReviewRead story
98

Claude Fable 5.1 Degraded: Status Page Marks Minor Incident Resolved

Anthropic's Claude status page has resolved a minor incident for Claude Fable 5.1 at 16:54:43 Asia/Shanghai (2026-11-08T08:54:43.667Z). Official impact window: 08:18–08:32 UTC = 16:18–16:32 Asia/Shanghai (elevated errors). Status updates: Investigating 16:28 → Monitoring 16:39 (fix deployed) → Resolved 16:54. Affected surfaces (now operational): claude.ai, Claude API, Claude Code, Claude Cowork. Overview: All Systems Operational. We haven't independently reproduced the errors or the recovery; facts below come from the status page / status API.

Claude Fable 5.1Read story
99

OpenAI Developers Plugin Adds a Showcase for Starter Projects

OpenAI Developers announced a new Showcase tab in its plugin, bringing together API starter projects for agents, creative tools, data visualization and voice. Projects can be opened in Codex for further development; available demo and source controls vary by project.

OpenAI Developers plugin/Codex/API demosRead story
100

OpenAI Monitors U.S. Business Mastercard Payment Issues After Mitigation

OpenAI says it has applied a mitigation for payment issues affecting some U.S. business Mastercards and is monitoring the affected services to confirm recovery. The incident remains monitoring, with impact none, not resolved. Earlier notices warned of blocked API credit purchases. DigIntoAI hasn’t tested payments.

API/billingRead story
101

Google Forms Actionable Insights Helps You Decide What to Do With Survey Results

A feedback form is easy to send. Turning its responses into a useful decision takes more work. You can read every comment and still finish with a vague list of things to improve.

GeminiRead story
102

When AI Remembers You It Also Needs to Know When You Have Changed

Last week, you asked an AI assistant to keep an email short. This week, you need a detailed report. It trims the explanation anyway, because it has remembered that you prefer concise writing.

AnthropicRead story
103

What It Takes to Turn ERNIE-5.5-Flash-Base Into a Useful Assistant

Baidu's ERNIE-5.5-Flash-Base offers a starting point for teams building their own language models. Turning the weights into a service involves choosing a task, preparing suitable data and measuring…

ERNIE-5.5-Flash-BaseRead story
104

Three Practical Tests for Ling-3.2-flash During Its Free Launch Period

Ant Ling launched Ling-3.2-flash on November 5 with a two-week introductory free period.[1] Developers evaluating multimodal APIs can use the trial to test chart reading, interface analysis and…

Ling-3.2-flashRead story
105

Grok Bot Degraded: Status Page Marks Minor Incident Resolved

Grok Bot’s minor incident was marked resolved at 23:15:07 on November 6, 2026 (Asia/Shanghai), with the update “This issue has been resolved.” The same incident opened at 21:57:57 as investigating. The overview now reads All systems operational, Grok Bot is available, and the active incidents list is empty. This was separate from the earlier computer-use issue. We haven’t tested this ourselves.

Grok BotRead story
106

Grok Bot Computer Use Failure in Some Regions Is Resolved

xAI reports that the Grok Bot computer use failure in some regions is resolved. The issue was marked resolved at 20:03:16 on November 6, 2026 (Asia/Shanghai). The official update says computer use in Grok Bot is working normally again for all users. The same incident had opened at 19:49:46 Asia/Shanghai as investigating / minor, when computer use failed in some regions while other Grok Bot features worked normally. Duration was about 13–14 minutes. We haven’t tested this ourselves.

Grok BotRead story
107

Cursor Lets You Build Live Apps with Cloud Agents and Share a Link

At 05:41:10 Asia/Shanghai on 2026-11-06 (2026-11-05T21:41:10Z), @cursor_ai announced Build Live Apps: a cloud agent can scaffold a web app, spin up Postgres when needed, place auth behind Cursor SSO, provision hosting, and return a shareable live link. Changelog date Nov 5, 2026. Docs require a cloud agent—not supported locally—on all paid plans with cloud agents enabled. Teams/Enterprise public sharing is off by default. Checks must pass and publish must be chosen before anything goes live. We haven't tested the flow.

Cursor/Cloud Agents/Build Live AppsRead story
108

Claude Rolls Out Personal Finance Beta to US Pro, Max, and Ultra Plans

At 02:00:35 Asia/Shanghai (2026-11-05T18:00:35Z), @claudeai introduced Personal Finance as a US beta. Connected accounts answer money questions in everyday language; access is read-only via Plaid—Claude cannot move, manage, or invest money. No ads or sponsored recommendations, no investment advice, and Claude does not sell its own cards, loans, or savings accounts. This week’s rollout covers Pro, Max, and Ultra in the US only—not Free, Team, or Enterprise—and the entry may not appear yet. We haven't tested the connection flow or balance accuracy.

Claude Personal Finance/ClaudeRead story