OpenAI's DevDay 2026 included more than 20 announcements, including the GPT-6 Astra model and updates to ChatGPT, Codex, and the APIs. The company also announced new security features and tools for developers.
OpenAI introduced dots, a proactive assistant that can keep working on complex projects and everyday tasks over time. OpenAI says the design keeps users in control while work moves forward.
Anthropic released Sonnet 5.5, its mid-tier model, which it says is 30 percent faster than Sonnet 5 and uses tokens at a much slower rate, and which beats Opus 5.5 on agentic coding benchmarks. Because Anthropic says its...
Basis reports that OpenAI's GPT-6 Astra finished a 50-tab tax workbook twice as fast as the earlier GPT-5.6 Sol model. The company says the new model's better grasp of user intent makes it more confident about using it i...
At its annual Connect event, Meta put its new personal AI agent Muse at the center of a consumer-focused AI push, while rivals OpenAI and Anthropic concentrate on enterprise and coding tools. On TechCrunch's Equity podca...
Meta's new AI app Muse has passed 3.4 million downloads since its September 8 launch, according to Sensor Tower, and has held the top spot on both the U.S. App Store and Google Play. At Meta Connect, the company announce...
Google has released Live Avatar for Gemini 3.8 Live, adding a near real-time video avatar with lip-sync and facial expressions to its conversational AI. It can call tools in the background while the conversation continue...
At its Connect event, Meta CEO Mark Zuckerberg presented Muse, the personal AI agent launched in early September and powered by the multimodal Muse Spark model, as the company's central consumer AI bet and a future 'pers...
OpenAI released updated GPT-6 Sol and Luna models, following the flagship GPT-6 Astra launch earlier in the month, positioning them as more efficient and accessible tiers for coding and high-volume clerical tasks. The co...
Anthropic released Opus 5.5, a model it calls its strongest to date, which outperforms the larger Fable model on many coding and knowledge-work benchmarks while costing less ($20/mTok output vs. $25) and running faster....
