Uploaded August 2025 | Updated September 2026, 2 weeks ago
OpenAI has launched two cutting-edge open-weight language models (gpt-oss-120b and gpt-oss-20b) that are comparable to proprietary models in performance while running efficiently on consumer hardware.
These models are optimized for reasoning, tool use, and real-world deployment, even on edge devices. In this video, I test their capabilities on my Windows laptop and on a Linux box with a 16GB NVIDIA GPU.
---
β Please support my channel on Patreon!
Get early access to videos, members-only content, behind-the-scenes updates, and join the Gary Explains Discord! Join here π patreon.com/GaryExplains π
Twitter: twitter.com/garyexplains
Instagram: instagram.com/garyexplains
#garyexplains
OpenAI has launched two cutting-edge open-weight language models (gpt-oss-120b and gpt-oss-20b) that are comparable to proprietary models in performance while running efficiently on consumer hardware.
These models are optimized for reasoning, tool use, and real-world deployment, even on edge devices. In this video, I test their capabilities on my Windows laptop and on a Linux box with a 16GB NVIDIA GPU.
---
β Please support my channel on Patreon!
Get early access to videos, members-only content, behind-the-scenes updates, and join the Gary Explains Discord! Join here π patreon.com/GaryExplains π
Twitter: twitter.com/garyexplains
Instagram: instagram.com/garyexplains
#garyexplains










