Meta launched Muse Spark Open, its largest U.S.-developed open-weight model featuring an 816B sparse Mixture-of-Experts architecture (42B active parameters), a 1M-token context window, and native Apple M5 Ultra optimization. Meanwhile, NVIDIA released Nemotron 3.5 Super VL, a 120B parameter MoE model that activates 12B parameters per token to deliver 5Ă— faster multimodal reasoning and throughput for enterprise agent workloads.
We’ll talk about:
High-capacity sparse MoE model optimized for Apple M5 Ultra Mac Studios, offering 1M-token context and leading open benchmark performance.
Multimodal open model achieving 600+ output tokens per second and 5Ă— higher inference throughput for enterprise agent pipelines.
Co-funding from OpenAI and Anthropic providing grants, compute, and AI tooling to support mathematical research.
New on-device security feature deploying Gemini to flag suspicious unknown-sender messages and potential fraud in real time.
Keywords: Muse Spark Open, Nemotron 3.5, Google Universal Cart, GPT-Live-1.1.
Podden och tillhörande omslagsbild på den här sidan tillhör
AIFire.co. Innehållet i podden är skapat av AIFire.co och inte av,
eller tillsammans med, Poddtoppen.
