Claude Opus 5 Launches: Same Price as Opus 4.8, Within 0.5% of Fable 5 on Coding Benchmarks
Anthropic launched Claude Opus 5 at $5/$25 per million tokens - Opus 4.8's price - while landing within 0.5% of Fable 5 on CursorBench at half the cost.
Breakthroughs in artificial intelligence, machine learning, and large language models - from research labs to real-world products.
Anthropic launched Claude Opus 5 at $5/$25 per million tokens - Opus 4.8's price - while landing within 0.5% of Fable 5 on CursorBench at half the cost.
Claude voice mode now runs on Opus and Sonnet for the first time, with connectors for Gmail, Slack, and Notion and support for 11 languages.
Health in ChatGPT launches today for US users. Connect Apple Health and medical records to give ChatGPT context on your actual lab results and medications.
Claude Record a Skill lets you teach Claude a task by walking through it on screen. Claude builds a repeatable automation from your recording.
Substack integrated Pangram AI detection, letting readers scan newsletters to see how much was written by a human. Writers can opt out or add an AI disclosure.
Google CodeMender is now in preview - an AI agent that scans code for vulnerabilities, proves they're exploitable, and generates tested patches automatically.
Meta is testing StoryKit, an AI app that generates personalized bedtime stories for children. Snap a toy, pick a lesson, get a story with music.
Claude Code now builds and tests iOS apps directly in Apple's Simulator - public beta, no extra permissions needed. Xcode required.
Google shipped three Flash-class Gemini models on Tuesday and dropped its first hint about Gemini 4. Gemini 3.5 Pro is still not out.
xAI launched Grok for Excel on July 20. Free Microsoft 365 add-in powered by Grok 4.5 - writes formulas, runs scenarios, and pulls from SharePoint and email.
Kimi K3 pricing: five tiers from free to $159/month annual, API at $3/$15 per million tokens, and when the 1M-token context unlocks.
Google is building a chip codenamed "Frozen v2" that bakes Gemini directly into hardware - 6-10x more efficient than its current AI chips, targeting 2028.