Burkov's TP Weekly #179: GPU rent vs buy, llm-d serving 753B model at 5-10x lower cost
burkov · x · 2026-09-25
Andriy Burkov released issue #179 of True Positive Weekly:
- An analysis of GPU renting vs buying economics
- An agentic data science war story
- llm-d: deploy a 753B open-weight model on NVIDIA H100s, serving thousands of concurrent coding agents at 5-10x lower cost
- An explainer on fine-tuning a 350M model for structured outputs in just 100 GRPO steps
- DeepSeek-V4.1-Flash released open-weight: supports images, strong at coding, 20x cheaper than equally capable closed competitors
- Plus an interactive Fourier transform intro, the Random Attention paper on KV cache eviction, and the 2026 Physics-Based Deep Learning book
More from Infra
- Parallax launches as AI energy startup with $117M from Founders Fund and others — Sethwinterroth · 2026-09-25
- New deep-dive article on scaling LLM inference in production — abhijithneil · 2026-09-25
- Lambda engineer shares local inference build rule: 27B models need 24-32GB VRAM — TheZachMueller · 2026-09-25
- Pokee AI demos 36B agent model running fully local on Snapdragon X2 Elite with 32GB RAM — Kyrannio · 2026-09-25
- AMD to present MXFP8 pretraining scaling on 1K+ MI355X GPUs at PyTorchCon 2026 — PyTorch · 2026-09-25
- AI energy startup Parallax launches with $117m from Founders Fund, Lux, Greylock and others — graceisford · 2026-09-25