2x NVIDIA RTX 5090 | GPT OSS 120B

Описание: In this video, we’re pushing AI hardware to the limits! We benchmark GPT OSS 120B running on two NVIDIA RTX 5090 GPUs in parallel, using shard-based loading across both GPUs. Watch as we test performance, latency, and scaling to see how this beast of a setup handles one of the largest open-source language models available today.

💻 What you’ll see in this video:

Dual RTX 5090 GPU setup in action

GPT OSS 120B model split into shards and loaded across both GPUs

Performance benchmarking and throughput tests

Insights into multi-GPU AI workloads

If you’re curious about AI model performance at extreme scales, this is a must-watch!

🔔 Don’t forget to like, subscribe, and hit the bell to catch more high-end AI benchmarking content.