PulseAugur
EN
LIVE 13:46:15

Lawyer's custom AI server with 12 Nvidia V100s runs DeepSeek V4.1 faster than DGX Spark

A lawyer named Nick Mermiges has significantly upgraded his private AI server, now equipped with 12 used Nvidia V100 GPUs, to run the DeepSeek V4.1 Flash model. This enhanced setup achieves speeds of up to 184 tokens per second for code and processes long documents much faster. Mermiges claims his custom server, built for $9,450, outperforms published DGX Spark configurations on single requests while costing substantially less. AI

IMPACT Demonstrates efficient deployment of frontier models on custom, cost-effective hardware, potentially inspiring similar setups.

RANK_REASON The cluster describes a user's custom hardware setup for running an AI model, not a release from a frontier lab or a major industry event.

Read on Tom's Hardware →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Lawyer's custom AI server with 12 Nvidia V100s runs DeepSeek V4.1 faster than DGX Spark

How we ranked this

Signal score
30 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The cluster describes a user's custom hardware setup for running an AI model, not a release from a frontier lab or a major industry event.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
infra, model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. Tom's Hardware TIER_1 English(EN) · Shane Downing ·

    Lawyer’s $9,450 private AI server of 12 used Nvidia V100s now runs every DeepSeek V4.1 Flash expert at 4-bit NVFP4

    A lawyer’s private AI server now runs DeepSeek V4.1 Flash on 12 used Nvidia V100 GPUs, with every expert at 4-bit NVFP4.