Home-built GPU rig claimed to run DeepSeek at 700 tokens/sec
philipvollet · x · 2026-11-11
A user showed off a home-built multi-GPU rack rig, claiming it delivers 700 tokens/sec for a "DeepSeek v4.1 Flash" setup. No config details or benchmarks were shared, and the model name is unverified — the post is mainly a local LLM deployment showcase.
More from Infra
- 5-person team maxes out all inference compute in 2 weeks, offers $20K referrals — ycombinator · 2026-11-12
- Akamai Bets on Inference Proximity: 20-Step Agent Chains Lose 4 Seconds to Distance — pdamodaran · 2026-11-12
- Akamai bets inference needs proximity: agent chains lose 4 seconds to distance alone — FinanceYF5 · 2026-11-12
- Reuters: Chinese military universities obtained restricted Nvidia AI chips despite US export controls — Polymarket · 2026-11-12
- Arm exec on the Agentic CPU: GPUs think, CPUs act — agents need their own class of silicon — BenBajarin · 2026-11-12
- Meta ships KernelEvolve, following Google's AlphaEvolve AI kernel-optimization play — RichardsonDx · 2026-11-12