Teknium 🪽

@teknium

Cofounder and Lead Engineer - Hermes Agent @NousResearch, prev @StabilityAI Github: http://github.com/teknium1 HuggingFace: http://huggingface.co/teknium

following
followers

Plans to turn Hermes agent into a DIY game dev studio

A user says they intend to make Hermes agent a do-it-yourself game development studio.

1 Source, 1d ago, first seen 1d ago

Hermes Agent catalog pairs 305 devices and tools with 1,547 project ideas

The Hermes Workshop says each entry explains how Hermes connects to the hardware and suggests use cases.

1 Source, 4d ago, first seen 4d ago

Hermes Index adds GPT 6 Sol and other models to its leaderboard

Nous Portal lists GPT 6 Sol fourth, with a 44.10 index score and $2.23 mean cost per task.

2 Sources, 2d ago, updated 2d ago

Nvidia's Nemotron 3.5 Super VL goes live on Together AI

Nvidia says the open model combines reasoning and tool use with text, image, document and video understanding.

45 Sources, 3d ago, updated 3d ago

Introducing NVIDIA Nemotron 3.5 Super VL

YouTube

A business reportedly runs on Nous Research’s Hermes

A user says they plan to show what Asher built on Hermes someday.

2 Sources, 6d ago, updated 6d ago

Claude Code 2.1.299 reportedly adds idle compaction

SemiAnalysis says sessions with 200k-plus context automatically compact after about 54 idle minutes.

7 Sources, 5d ago, updated 5d ago

A “give it a try” invitation links to a Tiny Fish post

A user shares the link with the message, “Give it a try!”

6 Sources, 4d ago, updated 4d ago

Fish Audio reportedly has an official Hermes Agent plugin

A post links to a Fish Audio plugin page in Hermes Agent’s documentation.

1 Source, 7d ago, first seen 7d ago

Asha and the coding-agent decompilation/game-mixing trend: lab, studio or funded startup?

A user sees Asha acting on the trend as a good sign and wonders what structure might fit.

8 Sources, 6d ago, updated 6d ago

Interrupt’s first sessions scheduled for October 13 at 10:30

LangChain listed an opening keynote by @hwchase17 and a fireside chat with Harrison Chase and Orange’s AI chief.

28 Sources, 2d ago, updated 2d ago

MISSINGNO. coding model was offered free for a limited time on Nous Portal in October 2026

NousResearch described it as highly token efficient and invited users to try it in Hermes Agent.

2 Sources, 10d ago, first seen 10d ago

Bringing back EverQuest with AI

A user suggested reviving EverQuest with AI, calling it “the best MMO of all.”

1 Source, 11d ago, first seen 11d ago

An enthusiastic reaction to a quoted post

A user calls the linked post “pretty sick” but gives no further detail.

1 Source, 12d ago, first seen 12d ago

Hermes is said to be the primary backend for a personal agent stable

A user says Hermes is their personal agent stable's primary backend and welcomes what they see as its continued growth.

9 Sources, 12d ago, updated 12d ago

Hermes Agent adds a separate model option for voice mode

The poster says a fast, reasoning-disabled voice model could shorten waits while the main agent handles harder jobs.

2 Sources, 14d ago, first seen 14d ago

Hermes Index debuts as a four-benchmark model leaderboard for Hermes Agent

Nous Research says the index aims to help users choose models and show labs how they perform in Hermes Agent.

5 Sources, 13d ago, updated 13d ago

OpenAI puts its Decisions API into public beta for developers

OpenAI says the Decisions API is now in public beta for all developers. In launch posts, the company said the GPT-6 Luna-powered API accepts text and image inputs, supports predicates, choices, and scores, and can be up to 10x faster than GPT-6 Luna through the Responses API.