Doop launched Sep 2, 2026 as an open-source, MCP-native, multiplayer AI design canvas. It splits from closed prompt-box tools on three axes: who runs the agent, who holds the memory, and who pays the token markup.
Meta shipped Muse Spark 1.3 with a pitch built on efficiency: roughly 20 percent fewer tool calls and 25 percent fewer tokens per coding task. Here is why tokens and tool calls per completed task, not raw benchmark score, are the metric that actually decides what an agent costs to run.
Adobe for Slack puts more than 70 creative tools inside chat conversations. The real story is describe-outcome-get-asset creation moving into the workplace surface, and why Adobe aimed it at enterprises, not solo creators.
Mercury 2.5 and Celeris-1 Magnus are diffusion LLMs that decode tokens in parallel to cut agent latency. Here is how they compare and when to swap.
Perplexity launched Hybrid Compute for its Computer agent on Apple silicon Macs, splitting a single task between cloud and local models so confidential files never leave the device.
Figma has turned its AI shaders from a static novelty into a working motion-and-code tool. A September 1 update adds motion, a code viewer, community publishing, and MCP export to its generative plugins and shaders.
Claude Mythos 5.1 is the same model as Fable 5.1 running with more permissive safeguards, available only through vetted cybersecurity and life sciences programmes.
Anthropic released Claude Fable 5.1 on September 1, 2026. Headline pricing did not move, but cache reads dropped from $1.00 to $0.25 per million tokens, a 75 percent cut worth up to 45 percent on agentic workloads.
Turn a product brief and a few reference images into a polished, multi-scene launch video in about ten minutes with AI, no After Effects required. A full step-by-step workflow.
fal.live is an infinite, interactive AI livestream powered by the autoregressive H3 Max Director model. Here is how continuous, steerable generation changes creative work versus the fixed-clip paradigm of Veo, Sora, Kling, and Runway.
Runway's Solaris generates interactive app interfaces in real time by rendering pixels instead of writing code, a genuinely different approach from AI code builders like Cursor and v0.
OpenClaw 2.0 shipped August 30, a full rewrite of the free, open-source AI agent platform you self-host with your own API key or a local model.
Tencent has open-sourced Hy4 preview, a 770B mixture-of-experts model with a 1M-token context window under Apache 2.0, aimed at coding, data analysis, and builder workflows.
A step-by-step ComfyUI workflow to turn one photo into a textured 3D model and a turntable video using the native TRELLIS2, Pixal3D, and SAM 3D Body nodes.
Gemini 3.5 Transcribe, GPT-4o Transcribe, Whisper, and Cohere Transcribe compared on accuracy, languages, streaming, diarization, and price so you can pick the right speech-to-text model.
Gemini Omni 1.1 Flash is the cheapest way to make controllable, long-form AI video, but not the highest-resolution. Here is how it compares to Veo, Sora, Runway, and Kling.
Adobe shipped a prompt-based AI Assisted Editor and Firefly Image 5 in Photoshop on August 27, 2026. Here is what is new and how to use it.
Cohere shipped Parse to general availability on August 27, 2026: a vision language model that turns documents into clean Markdown at $1.50 per 1,000 pages, topping the ParseBench leaderboard.
If you want an open-weights coding model you can route an agent to today or pull onto your own hardware, the honest answer in late 2026 is that it depends on what you own and what you trust.
The AI assistant you use every day now remembers you by default, and in 2026 that quietly became the feature that separates a chatbot from a working collaborator.
IBM shipped Granite 4.2 on August 25, 2026, a family of open, Apache 2.0 licensed agentic coding models in 3B, 8B, and 30B sizes that you can download and self-host today.
Alibaba's Wan 3.0 video model is now available natively inside ComfyUI with single-pass 30-second generation, up to 1080p output, and a 20-asset Omni-Reference system.
WebMCP lets your site hand AI agents named actions to call instead of scraping HTML. Here is how to expose your first WebMCP tool in about 30 minutes.
Turning a blog post into a video used to take days. In 2026 one person can do it in about an hour with four AI tools: an LLM to script, a voice generator, a text-to-video tool, and a free editor.