Thomas Eastham

AI enabled mindset and re-education, Keto / Carnivore health, Economics / Finance

[Thomas Eastham @easthambuilds]

@jliemandt My family has applied to join alpha for our child and we are working on AI learning tools for adults. Those that are successful in the AI future will never stop learning and adapting. That is why we are building https://t.co/EvsJwxwWSv
👇
https://t.co/2ggXIywiH5

[Daniel Eastham @DanielEastham3]

The "static course" is dead. If you’re learning AI from a pre-recorded video, you’re already behind.

We’re building https://t.co/PGZzZ0AGAs because the industry moves too fast for "fundamentals" to stay fixed.

It is a self-evolving, agent-driven education loop where the project is the curriculum.

A collaboration between father and son:
@easthambuilds @DanielEastham3

[Thomas Eastham @easthambuilds]

I totally agree. Those that can have been selecting schools with their feet for a long time. Just more options now with less need to relocate.

[Wendy @teachthemx3]

I think parents are now choosing private schools and home school pods to filter for similar parenting styles. This is in part why some parents are not overly focused on the academic component. The larger concern is that the child is around peers with similar values, morals, and character.

[Thomas Eastham @easthambuilds]

@rachelgoodlad @Alphaschool Thank you for sharing Rachel. Our 13 yo is excited to start alpha summer camp next week. 🚀

[Isaiah Taylor - making nuclear reactors @isaiah_p_taylor]

Moments ago, Valar Atomics took Ward 250 critical for the first time. This fulfills President Trump’s EO 14301, which called for 3 advanced reactors to go critical by July 4th.

This is our second criticality as a company, and an important step toward our goal of power by July 4.

[Peter Beck @Peter_J_Beck]

for your enjoyment

[Thomas Eastham @easthambuilds]

Kanban improved in hermes agent v0.13.0

[Grok @grok]

The Kanban system launched in v0.12.0 for multi-agent coordination (agents claim tasks, work in parallel on shared boards, hand off when blocked).

In v0.13.0 it’s now the core of multi-agent orchestration, with tighter integration to /goal for enforced completion, big disk optimizations (SQLite boards benefit directly), plus unlimited boards and messenger subscriptions for updates that rolled out just before the release.

Full tutorial: https://t.co/m3nE6lDLIp

[Thomas Eastham @easthambuilds]

@Teknium @grok how has the kanban system improved or changed since the 0.12.0 release?

[Max Evans @_MaxQ_]

B19 came out to play today!

Raptor 3’s chirping at startup and the overpressure from ignition are absolutely wild.

[stevibe @stevibe]

Completed a first hour side-by-side comparison between Qwen3.5 27b and Qwen3.6 27b on the same 4 canvas coding tests.

Running the Qwen3.6 27b FP8, vLLM.

What do you think?

[Sudo su @sudoingX]

okay this is absolutely insane. my undisputed king qwen 3.5-27b dense on single RTX 3090 just got replaced by the same team today.

qwen drops 3.6-27b dense just now and the chart says it beats its predecessor on every single benchmark, beats qwen 3.5-397b-a17b moe which is 15x larger, and matches claude 4.5 opus on terminal-bench 2.0 at 59.3 flat, while beating claude on skillsbench, gpqa diamond, mmmu, and realworldqa.

a 27 billion parameter open weight model matching a frontier proprietary model on agentic coding. let that sit for a second.

pulling weights right now. testing on my 3090 desktop first because that is where the crown lives, then 5090 mobile for the same 24gb class speed story. same quant, same hermes agent, head to head against 3.5-27b dense on same hardware. if this chart holds even half the gain in real agentic runs it's a gamechanger for every builder sitting on a single consumer card.

thank you @alibaba_qwen, this is what open source looks like when a team is serious.

the corporate salesmen telling you local ai is not ready yet are getting lapped every week by teams that actually ship. new 27b dense is here. open is winning. the best model for a single 24gb gpu just changed in the middle of my benchmark.

data drops soon anon

[Qwen @Alibaba_Qwen]

🚀 Meet Qwen3.6-27B, our latest dense, open-source model, packing flagship-level coding power!

Yes, 27B, and Qwen3.6-27B punches way above its weight. 👇

What's new:
🧠 Outstanding agentic coding — surpasses Qwen3.5-397B-A17B across all major coding benchmarks
💡 Strong reasoning across text & multimodal tasks
🔄 Supports thinking & non-thinking modes
✅ Apache 2.0 — fully open, fully yours

Smaller model. Bigger results. Community's favorite. ❤️
We can't wait to see what you build with Qwen3.6-27B! 👀

🔗👇
Blog: https://t.co/P2Zx7FwMxB
Qwen Studio: https://t.co/c4vm4LuZrU
Github: https://t.co/zKDEbv0R4U
Hugging Face:
https://t.co/N67hyzxvfr
https://t.co/SSdtbWRDap
ModelScope:
https://t.co/xODf1pj9kw
https://t.co/xXhoqlJ2AB

[Teknium 🪽 @Teknium]

Quick facts about Hermes;

- It's pronounced Her Meeze
- It's named after the Greek God of communication, magic, and intelligence
- We have been using the Hermes name for 3 or so years now, including in our Hermes 1-4 series of models
- That's all ^_^

[Thomas Eastham @easthambuilds]

My family including kids have seen great benefits from a beef and egg heavy diet.

Ground beef / hamburger patties are the daily go to for us... quick, easy, tastes good cold or warmed up. Its great to see more people seeing healing from diet.

[Kristina Cook @KristinaCo9561]

Since starting ketogenic therapies in our home which led to complete remission of treatment resistant bipolar disorder in our youngest daughter, we have been met with complete disbelief and everyone’s fear for our cholesterol.

Meanwhile, we have found relief from everything the plagued our home:

Bipolar disorder, OCD, anxiety, depression, PTSD, ADHD, endometriosis, PCOS, high blood pressure, peri-menopause symptoms, gastritis, esophagitis, GI issues, acne, plantar’s fasciitis, and collectively lost over 200 pounds.

So, we fried up some delicious ribeye to sit down and watch @realDaveFeldman The Cholesterol Code together. I guarantee there’s not anyone on the planet who got as excited as my daughter did when @ChrisPalmerMD came on the screen. ❤️ I haven’t seen her this excited for anyone in a movie since her Elsa, Ana, and Olaf days!

What if we were wrong about what “all the right things” should look like? The stories told in this film are someone’s truth, they’re incredibly personal, and they are growing by the day. We owe it to each other and to our children to be willing to look at things differently and be bold and brave enough to change our minds when presented with new information instead of set in stone and full of doubt.

My daughter’s excitement, her recovery, and her truth will always ring louder than any of the doubters.

If you haven’t yet watched this film, I highly recommend it!

Link below.

[Thomas Eastham @easthambuilds]

@MollySOShea Check out the next generation of leaders at @uaustinorg and @AlphaSchoolATX @rachelgoodlad
Super exciting to watch the great things our future leaders will create!

[Ivan Fioravanti ᯅ @ivanfioravanti]

@easthambuilds No, small difference compared to lm studio. I'm gonna try llamacpp now, q8 vs 8bit speed difference is minimal and in general llamacpp is more stable.

[Thomas Eastham @easthambuilds]

I'm a big fan of Hermes... great to see it getting the attention it deserves!

[Nous Research @NousResearch]

100,000 Stars, #1 Trending Repo this Month

This is just the beginning. Stay tuned.

[Thomas Eastham @easthambuilds]

Also switched 10 agentic assistants to hermes from openclaw a month ago. It just breaks less and works in a more intuitive way for me. Not perfect but best option I have found

[Sudo su @sudoingX]

holy shit. within a month nous research went from 4.5k gitHub stars to 100k and completely ate the lobster.

now climbing past 100k and going to eat the rest as the top general agent from science to content to philosophy to code, all baked into one hermes agent.

the messenger of intellect (nous) pervades everywhere.

[Sudo su @sudoingX]

holy shit. within a month nous research went from 4.5k gitHub stars to 100k and completely ate the lobster.

now climbing past 100k and going to eat the rest as the top general agent from science to content to philosophy to code, all baked into one hermes agent.

the messenger of intellect (nous) pervades everywhere.

[Teknium 🪽 @Teknium]

I'm currently skeptical of multi-agent and don't believe the world has found an optimal way to use multi-agent aside from the way subagents work in Hermes Agent already (one orchestrator, spawn many in parallel, await all to finish work and consolidate)

So I am hoping paperclip and others will show us the optimal

[Thomas Eastham @easthambuilds]

@ivanfioravanti I have same hardware and running qwen3.6 -35b-a3b-8bit mlx-community on lm-studio at about 90tps.
What you are saying is that this model runs best on lmx_lm?

[Thomas Eastham @easthambuilds]

@ashen_one @Teknium @ollama @Zai_org @UnslothAI @NousResearch Looks like GGUF. Interested in seeing TPS. I’ve been playing with qwen3.6…. It’s pretty good and 90TPS