NEWSFERENCE
THU, 22 Oct 2026 01:01:01
LIVE
$ today --liveF1TodayF2YesterdayF3ArchiveF4About
NEXT SCAN —
← BACK TO TODAY/CLUSTER · NOUS · AGENTS
CLUSTER · TIER 2
FIRST SEEN 4D AGO
NOUSAGENTS

Jensen Huang demos Hermes Agent at GTC Berlin

Jensen Huang closed his GTC Berlin 2026 keynote with a demo of Nous Research's Hermes Agent. The agent built a customized simulated training environment and trained a policy for Hugging Face's Microduck using NVIDIA hardware and robotics tools.

Sources
1
X mentions
57k ▲
First seen
4Dago
Velocity
+500%/6h
CONTRIBUTING SOURCES
2 ARTICLES
  1. X (Twitter)4D AGO
    x.com/nvidia/status/2112856720741740719
  2. X (Twitter)4D AGO
    x.com/NousResearch/status/2113010017645244623
X DISCOURSE
57k TOTAL · TOP 3
@NousResearch1H · 71.5K
Jensen closed his GTC Berlin 2026 keynote with this Hermes Agent demo as the finale. He introduced it by saying the way to build an agent is to ask an agent to build it for you, so he installed and set up a customized Hermes in one prompt. Then he used Hermes, running on @nvidia RTX Spark, to create a simulated training environment for the @huggingface Microduck. Hermes planned an obstacle course with Nemotron and had GPT-6 Astra build it using FreeCAD, Omniverse libraries, Blender, Newton, and Isaac Lab. Hermes trained thousands of simulated ducks on that course until the validated policy was ready to run on the real robot in the physical world. At Nous, we believe open source is a practical tool as much as a philosophical ideal. It gives users the ability to build powerful, complex workflows chaining numerous local+cloud models, devices, and tools themselves instead of waiting for a company to enable them.
@NousResearch5H · 68.9K
Jensen closed his GTC Berlin 2026 keynote with this Hermes Agent demo as the finale. He introduced it by saying the way to build an agent is to ask an agent to build it for you, so he installed and set up a customized Hermes in one prompt. Then he used Hermes, running on @nvidia RTX Spark, to create a simulated training environment for the @huggingface Microduck. Hermes planned an obstacle course with Nemotron and had GPT-6 Astra build it using FreeCAD, Omniverse libraries, Blender, Newton, and Isaac Lab. Hermes trained thousands of simulated ducks on that course until the validated policy was ready to run on the real robot in the physical world. At Nous, we believe open source is a practical tool as much as a philosophical ideal. It gives users the ability to build powerful, complex workflows chaining numerous local+cloud models, devices, and tools themselves instead of waiting for a company to enable them.
@NousResearch1D · 52.1K
Jensen closed his GTC Berlin 2026 keynote with this Hermes Agent demo as the finale. He introduced it by saying the way to build an agent is to ask an agent to build it for you, so he installed and set up a customized Hermes in one prompt. Then he used Hermes, running on @nvidia RTX Spark, to create a simulated training environment for the @huggingface Microduck. Hermes planned an obstacle course with Nemotron and had GPT-6 Astra build it using FreeCAD, Omniverse libraries, Blender, Newton, and Isaac Lab. Hermes trained thousands of simulated ducks on that course until the validated policy was ready to run on the real robot in the physical world. At Nous, we believe open source is a practical tool as much as a philosophical ideal. It gives users the ability to build powerful, complex workflows chaining numerous local+cloud models, devices, and tools themselves instead of waiting for a company to enable them.