Emanuel O.

Lead Data Scientist | 🌍 #Ergo

[Emanuel O. @inceptioncortex]

@chepurnoy @ErgoMinnow AEI! 🤖Sparking my interest in starting a side project w/ coworkers using Multi Agent Systems Human Feedback & Multi Models for content requests, pattern providers, bonds, evolving smart contracts.
Autonomous evolution w/ DAOs will be the next next digital economy! 🚀📈 #Ergo #AI

[Emanuel O. @inceptioncortex]

@Teknium @dotey

[Emanuel O. @inceptioncortex]

@Teknium @dotey Ah so on the right path! Discovered that skill earlier today but been using nvidia cosmos nano on a 3090 for image gen. Now also working on some custom themes for fun

[Emanuel O. @inceptioncortex]

@Teknium Curious how you generate these images and infographics? Looks amazing!

[Teknium 🪽 @Teknium]

Skill Bundles in Hermes Agent allow you to predefine a set of skills, then force load them all with one slash command.

Makes workflows that rely on a set of skills super easy!

Get access to them early with `hermes update`

[Alex Chepurnoy @chepurnoy]

Proper "web3" should be about Mesh networks, private chats (such as DarkIRC), Nostr/IPFS based services, p2p currencies using blockchain as global settlements and reserves layer. What "crypto industry" is trying to offer instead is hard to understand even.

[Emanuel O. @inceptioncortex]

This feature keeps giving!

[Teknium 🪽 @Teknium]

The Hermes Kanban now supports unlimited boards/projects, and you can now subscribe to updates from a project to your home channel across the gateway messengers you have setup!

Just run `hermes update` and you're good to go!

[Nous Research @NousResearch]

Hermes Agent now has multi-agent via the Kanban, new in v0.12.0.

Agents claim tasks from a board, work in parallel, and hand off when blocked. You watch progress and unblock from one easy view instead of juggling terminals.

We asked it to plan and make this video about itself:

[Emanuel O. @inceptioncortex]

@sudoingX Best combo for hermes-agent
> Dual rtx pro 6000-> Minimax 2.7 NVFP4 256k
> Single 3090 -> Qwen3.6 35b A3b 128k + Vision enabled

[Nous Research @NousResearch]

ComfyUI is the most flexible, composable, and powerful open-source media generation tool with a massive ecosystem of workflows and custom nodes.

Your Hermes Agent can now install, launch, manage, and run sophisticated @ComfyUI workflows on demand.

[elvis @omarsar0]

NEW paper from Alibaba.

A 30B MoE with only 3B active params matches Qwen3-235B on real tool-use workloads.

AgenticQwen-30B-A3B: 50.2 average on TAU-2 + BFCL-V4 Multi-Turn.

AgenticQwen-8B: 47.4.

Both more than double their vanilla Qwen baselines and close most of the gap to a 235B model.

How: two RL flywheels run in parallel.

- The reasoning loop mines the model's own errors into harder problems each round.

- The agentic loop grows simple linear tool-use trajectories into multi-branch behavior trees.

- Simulated users actively try to mislead the agent. The training distribution gets harder on its own.

Why it matters for agent devs: you can stop paying frontier prices for routine tool-use workloads.

And the flywheel recipe is reusable. Generate your hard examples from your own agent's failures, not from static synthetic data.

Paper: https://t.co/NGDXulumid

Learn to build effective AI agents in our academy: https://t.co/1e8RZKs4uX

[Andrej Karpathy @karpathy]

I like blockchain tech quite a bit because it extends open source to open source+state, a genuine/exciting innovation in computing paradigms. I'm just sad and struggle to get over it coming packaged with so much braindead bs (get rich quick pumps/dumps/scams/spams/memes etc.). Ew

[Emanuel O. @inceptioncortex]

@TheAhmadOsman At least it warms the pipes while agent delegates

[Emanuel O. @inceptioncortex]

This sentence should be pushed as the AI standard.

[Teknium 🪽 @Teknium]

Hermes Agent is model & tool backend agnostic for a reason, everyone should have access to AI. We don't dictate the rules of use for your agent, YOU do

[Emanuel O. @inceptioncortex]

@TheAhmadOsman Will try this to finally take the dust off my 3070

[Emanuel O. @inceptioncortex]

@sudoingX $ERG is my pick -- building open source and programmable money in Proof of Work that follows fundamentals and has grass roots community driven. There are already several community projects being built from agentic use.
https://t.co/ZlYnvcQXa7

[DomF ₳Σ @CryptoDomF]

Ergo is simple in its concept, but ridiculously powerful in capability 💪🏻

$ERG is one of the most under appreciated & undervalued L1s in crypto 😳

> Proof of Work + ASIC resistance
> Fixed supply + storage rent (demurrage)
> Secure by design, scalable via sub-blocks & NiPoPoWs
> Transparent L1 with optional privacy tools
> Smart contracts enabling a wide range of dApps
> Sigma chains for an ecosystem of connected chains
> Rosen bridge secured by Ergo for cross-chain value transfer
> Native oracles for reliable data feeds
> Decentralised stablecoins built on new monetary ideas

$ERG exists in a hostile environment for fair-launched projects, where tier 1 exchanges gatekeep liquidity, pick winners & push own narratives 👎🏻

Ergo is an evolution of Bitcoin. Same core DNA…. PoW, decentralisation, sound money, security… but with real programmability & scaling solutions Bitcoin will likely never adopt due to legacy design

If you believe in the Bitcoin vision, you should be watching Ergo closely 👌🏻

[Qwen @Alibaba_Qwen]

⚡ Meet Qwen3.6-35B-A3B：Now Open-Source！🚀🚀

A sparse MoE model, 35B total params, 3B active. Apache 2.0 license.

🔥 Agentic coding on par with models 10x its active size
📷 Strong multimodal perception and reasoning ability
🧠 Multimodal thinking + non-thinking modes

Efficient. Powerful. Versatile. Try it now👇

Blog：https://t.co/EXx5y466su
Qwen Studio：https://t.co/bg4tAU1p74
HuggingFace：https://t.co/w4pDX14DZS
ModelScope：https://t.co/SuRyLzdQiO
API（‘Qwen3.6-Flash’ on Model Studio）：Coming soon～ Stay tuned

[Emanuel O. @inceptioncortex]

@Teknium I find MoE better due to sub-agents or concurrency session scalling. But qwen is more stable overall.

[Emanuel O. @inceptioncortex]

@Teknium I set a daily cronjob to digest top 10 posts on /r/locallama. Does a quick summary and top comment discussions. All in the palm of a telegram chat.

> Bonus: can ask for more details or similar topics.

[Emanuel O. @inceptioncortex]

@DODOREACH @NousResearch @Teknium Yes! So much potential.. agents can represent "feelings" or states of mind