aipingnews.com · Sun 15 Nov

Sunday 15 November

Baseten publishes DeepSeek engine, reports 2,060 tokens per second on eight B300s

AI-generated summary Baseten published an inference engine for the DeepSeek V4.1 Flash model, reporting a speed of 2,060 output tokens per second on eight Nvidia B300 GPUs. Teams with compatible hardware can run the DeepSeek weights on the engine, which was written in Rust and CUDA and released under the Elastic License 2.0.

Earlier:

Perplexity brings Teammates for recurring work to its Mac app

AI-generated summary Perplexity said that its Teammates AI agents are available to Max subscribers in the company Mac app for recurring tasks. Perplexity says subscribers can assign work that continues across several sessions and check progress in a dedicated conversation.

Also covered byposts.design
Earlier:

Kitesurf: A browser built for software

AI-generated summary Cloudflare has made its stateless browser engine for AI agents open source under the Apache 2.0 license, running entirely on Workers with no browser container. Cloudflare describes the release as experimental, noting that real-world use is currently limited to early testers because there is no video or WebGL support and no built-in authentication.

Also covered byAI PingDotsBot
Earlier: