Yacine's 90-minute deep dive: latent MoE, aggressive GQA inside Nvidia's open model
yacinelearning · x · 2026-10-01
Yacine released a 1h30 discussion with @llmwizard digging into the inner workings of NVIDIA's frontier-class open model.
Topics covered include latent MoE, speed optimizations, aggressive GQA configurations, and what he calls "bonkers" linearization of attention. The recurring thesis: a faster model is a smarter model — inference speed itself is capability.
More from Infra
- Mobile AI Factory: agents auto-build on-device SLMs with 19x less GPU time — pmttyji · 2026-11-16
- vLLM serves DeepSeek V4.1-Flash at 954K tokens/s per GPU, $38.96/GPU-hour on GB300 NVL72 — bookwormengr · 2026-11-16
- Colossus 2's 1.2M B200s could generate ~$400B/year in tokens at DeepSeek Flash rates — bookwormengr · 2026-11-16
- BIS issues rule blocking remote access to US AI chips via third-country data centers — teortaxesTex · 2026-11-16
- Qdrant 1.20.1: 50% fewer cold-storage reads, ~2x faster quantization, agent-ready setup — qdrant_engine · 2026-11-16
- One 16GB GPU runs songs, voice, images, 720p video and local RAG — all offline and open source — llama-of-death · 2026-11-16