In an informal SVG generation benchmark, Alibaba's Qwen3.6-35B-A3B model running locally via a 20.9GB quantized version outperformed Anthropic's newly released Claude Opus 4.7. The test, which asked models to generate SVG illustrations of pelicans and flamingos on bicycles, showed the smaller local model producing more accurate bicycle frames and more creative outputs.
OpenAI's Codex Desktop can now control your computer, run background automations, and includes an in-app browser with click-to-select elements. The update adds automation memory across sessions and access to over 100 curated plugins, though the computer control feature is MacOS-only and unavailable in the EU.
Google is integrating AI Mode directly into Chrome's interface across desktop and mobile. The update adds inline prompts on the New Tab Page, a side panel for conversations alongside web browsing, and address bar access with file upload capabilities.
Google rolled out split-screen browsing for AI Mode on Chrome desktop, letting users view webpages alongside conversational search results. The update also adds multi-tab context search across Chrome desktop and mobile.
Google has integrated its Nano Banana image generation system with Gemini's Personal Intelligence feature, enabling the AI to create images informed by user data from Gmail, Photos, Calendar, Drive, and other Google apps. The feature rolls out to Plus, Pro, and Ultra subscribers in the US first, with Europe excluded from the initial launch.
Anthropic released Claude Opus 4.7 with improved coding capabilities, higher-resolution vision, and a new reasoning level. The company publicly acknowledged the model underperforms its unreleased Mythos system, which remains restricted due to safety concerns.
Perplexity has launched Personal Computer, a Mac-based AI assistant that integrates with local files, native apps, and browsers. The feature is available to Perplexity Max subscribers ($200/month) and waitlist users starting April 16, 2026.
Roblox is deploying agentic features to its Assistant tool that plan, build, and test games through multi-step workflows. The enhanced Planning Mode analyzes code, asks clarifying questions, and creates editable action plans before implementation, while new AI-driven playtesting tools automatically identify and fix bugs.
Google's Gemini can now generate personalized images by pulling data from users' Google Photos libraries through its Personal Intelligence feature. The integration uses Google Photos labels to identify people and objects, then generates images via the Nano Banana 2 model that reflect users' tastes and lifestyle.
Anthropic has released Claude Opus 4.7, its most powerful generally available model, though it scores lower than the company's Mythos Preview model on every evaluation. The company intentionally reduced Opus 4.7's cybersecurity capabilities during training as it tests safety measures before releasing more powerful models.
Anthropic has released Claude Opus 4.7, its latest generally available AI model focused on advanced software engineering. The model shows improvements in handling complex coding tasks with less supervision, enhanced vision capabilities, and better instruction following, while introducing a new tokenizer that increases token usage by 1.0-1.35× depending on content type.
Anthropic has released Claude Opus 4.7, the latest version of its flagship Opus family designed for long-running, asynchronous agent tasks. The model features a 1 million token context window and costs $5 per million input tokens and $25 per million output tokens.
Anthropic released Claude Opus 4.7, a new model that the company says is 'broadly less capable' than its most powerful offering, Claude Mythos Preview. The model includes automated safeguards that detect and block prohibited or high-risk cybersecurity requests.
Anthropic released Python SDK v0.96.0 on April 16, 2026, adding support for Claude Opus 4-7, a new flagship model. The update introduces token budgets for cost management and user profiles for personalized interactions.
Alibaba has released Qwen3.6-35B-A3B, the first open-weight model in the Qwen3.6 series. The model features 35B total parameters with 3B activated, a native 262K context window extensible to 1.01M tokens, and achieves 73.4% on SWE-bench Verified using 256 experts with 8 activated per token.
Character.AI has launched Books mode, a structured roleplay feature that lets users interact with over 20 classic public domain titles including Alice in Wonderland, Pride and Prejudice, and Dracula. The feature includes book arc mode that follows original narratives and off-script mode for free interaction.
Canva released AI 2.0 today, rolling out to the first one million users who access its homepage. The update introduces a unified conversational interface that allows users to generate and edit designs through text prompts, with the platform claiming users can create "multi-channel campaign plans" and execute them entirely through natural language descriptions.
Canva has released Canva AI 2.0, an agentic assistant that automatically calls design tools based on text prompts and creates editable layered designs. The update includes integrations with Slack, Gmail, Google Drive, Calendar, and Zoom for context building, plus web research and task scheduling capabilities.
OpenAI has added native sandbox execution and governance controls to its Agents SDK, allowing enterprises to deploy AI agents with isolated compute environments and credential separation. The SDK now supports major cloud storage providers including AWS S3, Azure Blob Storage, Google Cloud Storage, and Cloudflare R2, with built-in integrations for sandbox providers like E2B, Modal, Blaxel, and Vercel.
Opera is extending its MCP-compatible browser connector feature to Opera One and Opera GX, making it available beyond the subscription-based Opera Neon browser. The feature allows AI assistants including Claude, ChatGPT, and Gemini to access browser tabs, read page content, and take screenshots during live browsing sessions.