224B (12B active), 1M context, natively multimodal. Fast, efficient and built for everyday productivity.

Isomorphic Labs announces it has raised $5 Billion in Series C funding. The round was co-led by Greenoaks and Sequoia Capital, alongside Andreessen Horowitz (a16z) and several other new investors, with continued significant investment from Alphabet.

Suno's tool turns songs into lyric videos and album art

9–15 November 2026
Anthropic, PBC opens roadshow for Nasdaq listing under symbol ANT
Meta's smallest model targets repeatable agent tasks at low cost
Teammates come to the Perplexity Mac app for Max subscribers
Specialist security model matches GPT 6 Astra at under 2% cost
SenseTime's 9B model reuses layers twice to reach 27B-class results
Conditional DNA model predicts variant effects in specific cell types
Claude joins Google Meet calls in beta on Team and Enterprise plans
OpenAI introduces Daybreak cybersecurity products under a Codex plugin
Whiteboards and Sheets in beta, Dashboards now Claude Data
Alibaba's Qwen4-Max model with 1M-token context and lower API prices
Today at Microsoft Ignite 2026, we’re announcing the general availability of Azure Cobalt 200-based virtual machines (VMs) . Cobalt 200 VMs deliver up to 50% better performance than Cobalt 100 and are now available in 14 regions, expanding to 23 regions by the end of 2026.
Seven models pitched at rivals' quality for a fraction of the price
Two coding models tuned for everyday enterprise tasks
As software developers, we’ve always wanted to embed intelligence inside our software, applications, and workflows, rather than anticipate all scenarios and hardcode every decision.
Open-source agent framework reaches general availability after June preview
New decision models add audio input and five tasks
OpenAI says GPT-6.2 Astra and Sol are now available
New image models and Canvas workspace in public preview
Hugging Face switches Inference Providers to OpenRouter's router
New Vera Rubin GPU targets simulation for science and engineering
£1m donation to help 25,000 young people build AI skills
Krea Video gains a workflow for swapping characters and backgrounds
SageMaker adds auto-labeling, synthetic data and quality checks
OpenTSLM time-series language models deployable in AWS environments
Remote access, dictation and browser use come to JetBrains IDEs
Inference-time scaling expands latent capacity for 3D shape modeling
Video: SenseTime
Image from microsoft.ai
Video: OpenAI
Image from elevenlabs.io
Image from about.fb.com
Image from replit.com
Image from legora.com
Image from harvey.ai
Image from cohere.com
a16z Podcast· 17 Nov