@tedmurphy profile picture
Ted Murphy @tedmurphy

CEO of COEY. Founder of NASDAQ:IZEA. Influencer Marketing Pioneer. AI Automator of all the things.

Joined January 2007
  • Tweets 20,635
  • Following 702
  • Followers 28,438
1,317 Photos and videos
Real signal from @Microsoft Image APIs matter more than demos coey.com/resources/blog/2026…

Microsoft’s MAI-Image-2 Gets Serious About Real Work - COEY

Microsoft has rolled out MAI-Image-2 in Microsoft Foundry, and this launch matters for a very non-glamorous reason: it looks built for work, not just for showing off in a demo thread full of “cinem...

coey.com
1
65
Voice got real when @OpenAI made it programmable coey.com/resources/blog/2026…

OpenAI’s GPT-Realtime-2 Push Makes Voice Agents More Operational - COEY

OpenAI has introduced GPT-Realtime-2 alongside GPT-Realtime-Translate and GPT-Realtime-Whisper, expanding its Realtime API into something much closer to actual voice infrastructure than a shiny...

coey.com
76
Big context gets real when @xai hits the API coey.com/resources/blog/2026…

xAI Grok 4.3 Pushes Into Long-Context Ops With 1M Tokens and API Access - COEY

xAI has introduced Grok 4.3 through its developer platform, and the headline feature is hard to miss: a 1,000,000-token context window paired with API access, tool calling, and pricing that looks...

coey.com
64
Why @Google folding video into Gemini matters for marketers coey.com/resources/blog/2026…

Google’s Gemini “Omni” Leak Signals Video Is Moving Into the Assistant Layer - COEY

Google appears to be testing built in video generation inside Gemini, with leaked interface screenshots showing a new video tab labeled “Powered by Omni.” If that holds, this is more than one more...

coey.com
70
Workflow speed matters more when @OpenAI cuts weird moments. coey.com/resources/blog/2026…

OpenAI’s GPT-5.5 Instant Is Here, and the Real Upgrade Is Workflow Speed With Fewer Weird Moments -...

OpenAI has rolled out GPT-5.5 Instant as the new default Instant model in ChatGPT, replacing GPT-5.3 Instant and sharpening the company’s pitch around faster everyday AI that is also more reliable,...

coey.com
1
55
2M tokens matters when @Google cuts workflow duct tape coey.com/resources/blog/2026…

Google Pushes Gemini 3.1 Ultra to 2M Tokens, and That Changes the Workflow Math - COEY

Google’s Gemini API appears to be getting a much louder flagship story with Gemini 3.1 Ultra: a multimodal model described in recent Google-linked and developer reporting as supporting up to a 2...

coey.com
1
62
Open weights matter when @MistralAI meets workflow and @coeycorp coey.com/resources/blog/2026…

Mistral Medium 3.5 Wants to Be the Open Agent Model Adults Can Actually Deploy - COEY

Mistral has launched Mistral Medium 3.5 alongside new agent features in Le Chat and remote agents in Vibe, and this one lands with a much more practical ambition than the usual “look, our chatbot is...

coey.com
73
Cheaper video changes the math. @GoogleCloud just made draft first AI video workflows a lot easier for brand teams. coey.com/resources/blog/2026…

Google’s Veo 3.1 Lite hits Vertex AI - COEY

Google has brought Veo 3.1 Lite to Vertex AI, adding a lower-cost video generation tier aimed at teams that need volume, not just wow-factor. That distinction matters. Generative video has spent...

coey.com
91
Multimodal gets useful when @nvidia makes it callable. One model reading decks, audio, video, and docs is less demo theater, more workflow infrastructure for brand teams. coey.com/resources/blog/2026…

NVIDIA Nemotron 3 Nano Omni Makes Multimodal AI More Operational - COEY

NVIDIA has launched Nemotron 3 Nano Omni, an open multimodal reasoning model built to process text, images, audio, and video in one system instead of forcing teams to stitch together a tiny AI...

coey.com
1
69
Demand for semiconductors and memory is not going to slow down any time soon. Before the data center super cycle ends, the robot super cycle will begin, which will in turn drive another data center super cycle.
95
Multimodal gets useful when it plugs into systems. @nvidia @coeycorp coey.com/resources/blog/2026…

NVIDIA’s Nemotron-3-Nano-Omni Makes Multimodal AI More Operational - COEY

NVIDIA has launched Nemotron 3 Nano Omni, an open multimodal reasoning model designed to handle text, images, audio, and video inside one system instead of forcing teams to stitch together a little...

coey.com
61
AI video gets practical when audio ships too. @alibaba_cloud via @coeycorp coey.com/resources/blog/2026…

Alibaba’s HappyHorse 1.0 Makes AI Video More Workflow-Ready - COEY

Alibaba has introduced HappyHorse 1.0, a new video generation model that is getting attention for one reason the market actually cares about: it can generate video and synchronized audio together...

coey.com
75
Meta put AI where buyers lose time. @Meta got practical. coey.com/resources/blog/2026…

Meta’s AI Business Assistant Goes Global - COEY

Meta has expanded its AI Business Assistant to advertisers and agencies across major global markets, placing a conversational support and optimization layer directly inside Ads Manager and Business...

coey.com
68
AI images get practical when typography works. @alibaba_cloud coey.com/resources/blog/2026…

Alibaba’s Qwen-Image-2.0-Pro Wants to Make AI Typography Actually Usable - COEY

Alibaba has introduced Qwen-Image-2.0-Pro, a higher-end version of its Qwen image model built for the kind of work most generators still fumble: readable text, structured layouts, and visuals that...

coey.com
82
DeepSeek got interesting when @deepseek_ai made 1M context deployable coey.com/resources/blog/2026…

DeepSeek V4 Brings 1M-Context Open Weights Into the Automation Race - COEY

DeepSeek has entered the long-context arms race with new V4 open-weight releases that pair a 1,000,000-token context window with Mixture-of-Experts efficiency and commercially usable MIT-licensed...

coey.com
66
GPT 5.5 gets useful when @OpenAI meets APIs. @coeycorp coey.com/resources/blog/2026…

GPT-5.5 Pushes OpenAI Deeper Into Real Agent Work - COEY

OpenAI has introduced GPT-5.5 and GPT-5.5 Pro, and the important part is not that the benchmark charts got another glow-up. It is that OpenAI is pushing harder into AI that can plan, execute, and...

coey.com
65
Native 4K matters when @kling_ai output has to survive edits. coey.com/resources/blog/2026…

Kling AI Brings Native 4K to Video 3.0 - COEY

Kling AI has added native 4K generation to Kling Video 3.0, pushing its video stack closer to something brands can actually ship instead of just admire in a demo thread. The update is live through...

coey.com
113
Why @Kimi_Moonshot K2.6 matters real agent work, not demo bait coey.com/resources/blog/2026…

Moonshot AI’s Kimi K2.6 Pushes Open Models Closer to Real Agent Work - COEY

Moonshot AI has released Kimi K2.6, and this is the kind of model launch that matters for a very simple reason: it looks less like “another chatbot with a new benchmark chart” and more like infrast...

coey.com
65
Why @OpenAI Images 2.0 matters for marketers coey.com/resources/blog/2026…

ChatGPT Images 2.0 Pushes AI Visuals Closer to Real Workflow Territory - COEY

OpenAI has launched ChatGPT Images 2.0, a major refresh to image generation and editing inside ChatGPT that looks less like a novelty upgrade and more like a serious attempt to make AI visuals useful...

coey.com
78
OpenAI @OpenAI is making agents easier to govern coey.com/resources/blog/2026…

OpenAI’s Agents SDK Gets Sandboxed Workspaces, and That’s a Big Deal for Real Automation - COEY

OpenAI has expanded its Agents SDK with native sandbox support and workspace-style execution controls, a practical upgrade that moves agent automation a little further out of demo-land and closer to...

coey.com
87
Load more