Alibaba's Qwen team has released Qwen-Image-2.1-Turbo, a faster checkpoint of its open-weight Qwen-Image-2.1. It generates and edits images in 8 denoising steps instead of 40, on the same 7B architecture, and is also offered through a hosted API.
OpenAI's Decisions API, running on GPT-6 Luna, is now in public beta. It returns typed probabilities, choices and scores about ten times faster than the Responses API, at $0.10 per million input tokens with no output charge.
TypeSafe, maker of the Jev AI model, is valued at $7.5 billion only weeks after launch. The company claims Jev runs much faster and uses far fewer tokens than large language models.
Cloudflare is adding Clef-omni to its Clef family of decision models; it handles audio, video, images and text in a single pipeline. The company has also cut Clef-flash prices and made Clef inference up to twice as fast.
SiliconANGLE's weekly column says it is too early to call an AI bubble, given the money still flowing into AI companies. It notes OpenAI told investors its revenue was $18 billion below the figure previously cited, alongside a busy week of model releases and a Trump crackdown on tech green cards.
Nace.AI has open-sourced Drex 1.5, a 9-billion-parameter decision model that assigns a probability to every option in a single forward pass. It scores 58.08 on Decision Index 0.3.1 and handles up to 128,000 tokens of input.