The open‑source inference engine upgrades to version 0.5.22, boosting throughput for several models and adding native Mac support.
NoteBy OossaPublished by Oossa: 1 min read
The SGLang project released version 0.5.22. It speeds up DeepSeek‑V4.1 by up to 2.3× on NVIDIA H200 GPUs and adds AMD MI350X/MI355X support with up to 1.9× throughput. New models such as EmbeddingGemma 2 and Kandinsky 6 are now included. The engine can now run on Apple Silicon Macs using the standard PyTorch backend. Upgrading requires Python 3.11 or newer.
Developers can run larger LLMs faster on more hardware, including Macs, without buying new servers.
Models
Canva's new AI creates up to 2K images with detailed layouts and editable text, available to all users via Dream Lab and the API.
Models
The update adds 14 voices, including Hindi and Finnish, and reaches existing generative-voice users without code changes or a price increase.
Models
Uber says its global workforce will use ChatGPT Work after an earlier 1,000-person pilot. The company has not disclosed the rollout’s cost.
Oossa · Newsletter
The week in AI, explained
Every Monday: the stories worth knowing, in plain language. Free, no spam.
