All sections
News
507 stories · page 3 of 57
InfraAMD completes its World Labs acquisition, Fei-Fei Li joins as chief scientistAMD said on 16 November 2026 that its acquisition of World Labs is complete. Fei-Fei Li becomes executive vice president and chief scientist, reporting directly to Lisa Su.InfraNVIDIA swept every MLPerf Training v6.1 test, and Vera Rubin debuts with leading numbersNVIDIA said on 16 November it swept every MLPerf Training v6.1 benchmark and made its first MLPerf Endpoints submission. According to NVIDIA, Vera Rubin NVL72 delivers up to 2x more performance per GPU than GB300 NVL72, and the Endpoints results are still pending MLCommons verification.FrontierVercel adds Qwen4-Max to AI Gateway: the first model in the Qwen4 series, with native vision and a 1M-token contextVercel announced on 15 November that Qwen4-Max from Alibaba is available on AI Gateway. According to Vercel, it is the first model in the Qwen4 series: a reasoning model with native vision and a 1M-token context window, for coding, professional work and long-running agent tasks.SecurityUnrealIRCd: a hole in labeled-response lets any user inject server commands, and CISA added it to the catalogA vulnerability in UnrealIRCd's labeled-response module: the "label" tag is echoed into replies without escaping, and any user on a network with two linked servers can inject commands with server authority. CISA added CVE-2026-114713 to its catalog of known exploited vulnerabilities on 13 November, with a due date of 16 November.BuildersVercel Token Pass now gives $150 in tokens instead of $80, for the same $30 per team per monthVercel raised the Token Pass credit from $80 to $150 of tokens per month at the same $30 per team. The add-on is in Early Access for Pro teams, with limited spots and a waitlist.Buildersllama.cpp and ggml reach 1.0.0: a broken API will now show up in the major versionOn 13 November ggml-org released libllama and libggml 1.0.0. From here on, every incompatibility in the public C API and in the ABI bumps the major version, so anyone who built a local model into an app can read the number as a warning.BuildersDevin Apps: you describe the internal tool, Devin builds it and puts it on a private address, and the bill runs while the machine is awakeOn 13 November Cognition launched Devin Apps: you describe the tool, Devin builds it and puts it at an address you reach only by signing in to Devin. Billing follows how long the machine is awake. The announcement gives no prices, and by Cognition's own words the apps are not for high-traffic services.BuildersCloudflare can now pause Workers on its own when you hit your cap: Spend Management enters betaOn 13 November Cloudflare released Spend Management in beta for pay-as-you-go customers: you set a spending cap and, when it is reached, the product is paused automatically. For now it covers Workers and Durable Objects.
507 stories