Computing & Hardware
Lawyer builds private AI server with eight used Nvidia V100 GPUs
A lawyer runs DeepSeek V4.1 Flash on eight used 2018-era Nvidia V100 GPUs at up to 215 tokens/s, keeping case files in the office.
A lawyer runs DeepSeek V4.1 Flash on eight used 2018-era Nvidia V100 GPUs at up to 215 tokens/s, keeping case files in the office.
This is a summary of the story. Read the full article at Tom's Hardware (Computing & Hardware) ↗