OpenAI launches GPT-6.1 Sol at one-fifth the price of its flagship Astra model
“near-Astra intelligence for coding, computer use, and professional work at one-fifth of Astra's standard API input and output token prices”
computer-use
12 tracked signals on computer-use.
Google DeepMind adds computer use capabilities to its Gemini 3.5 Flash model.
Windows 365 for Agents is now generally available, giving AI agents their own cloud PCs to operate Windows apps like a human.
“You need to give your agent a computer. That's what Windows 365 for agents does.”
OpenAI and Ironclad are training AI agents on complex legal contracting workflows
OpenAI's computer use agents are now '180 degrees different,' approaching superhuman software operation speed
“180 degrees different”
Standard computer use benchmarks are gameable by blind replay scripts, invalidating frontier model comparisons
“if you try to evaluate this kind of agent on standard benchmarks such as OSWorld or MobileWorld, you will see that the success rate of this agent compared to the frontier model from which the agent was extracted is actually the same or even better”
Computer-use models, not APIs, will agentify the long tail of the web
“this last bit that suddenly the web will provide you APIs for accessing things is just delusional”
Agent reliability, not capability, is the critical unsolved problem for enterprise automation trust.
“if your agent one in four times deletes a database, you will never touch that agent again”
Daytona enables AI agents with cloud computers accessible via API, signaling end of localhost development
“The end of localhost”
Web agents face a model capabilities overhang as infrastructure, not models, is now the bottleneck
“there's a huge model capabilities overhang in this category specifically, and you all here can hopefully solve it”
AI practitioner guides have shifted focus from chat interfaces to agentic systems doing hours of real work
“The most powerful way to use AI is to give it access to your computer. You do that by downloading the ChatGPT or Claude apps and picking a mode to use.”
Hugging Face announces Holo4, a generalist computer-use agent system