ST

Project Manager-Vibe Coder-

[ST @SethTam]

@devinfoley Way to go devin amazing product!

[ST @SethTam]

@lmstudio Nice work @lmstudio always making your products better and better the new Gemma 4 12b is already working and it just was released yesterday!

[LM Studio @lmstudio]

Meet LM Studio's mobile app.

Your local models, now in your pocket.

[Kalshi @Kalshi]

JUST IN: SEC officially eliminates $25,000 minimum for day trading

[ST @SethTam]

@KyleHessling1 This model is the fastest of all the Qwen3.5-9b work you have created Kyle insane work! Thanks!!!

[ST @SethTam]

@KyleHessling1 It is twice as fast but I am always learning when it comes to making these models work better!

[ST @SethTam]

@KyleHessling1 The performance of Qwen3.5 9B DeepSeek v4 Flash is double what I get on other 9B models you have supercharged! Thanks 40 tok/sec

[ST @SethTam]

@hugobowne @gdb @HamelHusain OMG Greg Brockman Watches This show how Epic!

[Katia Ameri 🗽 #NYTechWeek @KatiaAmeri]

Boston —> NY Tech Week, here we go!🗽

No rest for the weary. Spending the day clearing up the inbox and getting ready for round two. 🙃

[WorldofAI @intheworldofai]

🚨BREAKING: OpenAI's Codex is now completely FREE to run locally with Ollama.

No API costs. No rate limits. 100% private on your machine.

You can now use both the Codex App and Codex CLI with powerful open-source models like DeepSeek V4, Gemma 4, and Qwen 3.6.

Here's how to set it up in minutes:

[ST @SethTam]

@bindureddy How is the pricing on this model I noticed it is live now in @abacusai

[ST @SethTam]

@TrustVanta Without a doubt @yayalexisgay is so good at corporate humor!

[ST @SethTam]

@KingBootoshi Listen to this man he has a robot and is deep into Agentic Engineering as well!

[ST @SethTam]

@KyleHessling1 A program I am working on was going really slow using qwen3.5 35b my hardware is 8gb VRAM so too big for what I have switched to Jackrong/Qwen3.5-9B-DeepSeek-V4-Flash and it runs so fast now 40 tok/sec THANK YOU KYLE !!!!

[ST @SethTam]

@iBrews Any chance you could make a how to video on how you use this. Why take it down. We need cool tools. Great work as always thanks

[ST @SethTam]

@LottoLabs What is the spec of your system. VRAM, sys ram, and CPU?

[Kyle Hessling @KyleHessling1]

BREAKING! Qwopus 3.6 27B is LIVE!

Thank you for your patience on this one, but I believe you'll find the wait was worth it!

We've benchmarked this thing up and down, verified that it holds at least a 75.25% (152/202) in the initial 202 SWE bench solves. Not a full run of 500, but it shows the agentic coding quality from the original 27B is retained while adding all of the additional Qwopus benefits across many domains. As always, Jackrong is absolutely cooking here!

COT quality has improved significantly through the inversion techniques from our Negentropy proof of concept. It also went through thorough curriculum training. You can check out the MMLU pro benchmarks on the model card, but it improved a whopping 10 points over the base model in physics, as well as meaningful jumps in Chemistry, business, and computer science.

However, the best part is that I was able to build an entire survival shooter game using this local model entirely. I genuinely was blown away by the results, which you can play right now on my HF space (link in comments below). "Qwopus Commander" was completed in 9 turns of Qwopus 3.6! To test the new long context training, I made it re-output the entire 3000+ line program each turn, and it would make fixes and add features that I requested in large prompts, while perfectly replicating the entire rest of the game from context. What's more is that I did it all at Q8 KV cache quantization, and never had an issue over the entire 303k token run!

IMPORTANT: Run it at --temp 0.75 to 1. Mess with it in that range for your use case. Higher temp actually lets the fine-tune shine and be exploratory and is also more stable. Swe Bench was run at temp 1, the game was built mostly at 0.8!

We're so blessed to have all of you here and using the models! The support means so much! Please let me know what you build with it in the comments! Or if you have any issues getting it up and running, I will try my best to get back to you!

Looking forward to seeing what you legends produce with it this weekend!

https://t.co/AEl3APtTLk

[ST @SethTam]

@KyleHessling1 I have successfully hooked up claude code CLI via LM Studio with Qwopus3.5-9B-Coder-Exp and it works perfectly and it is so fast 38 tok/sec on 8gb VRAM Nvidia Blackwell RTX 2000. THANK YOU I needed this extra speed!

[ST @SethTam]

@KyleHessling1 @NousResearch Great work KYLE....loaded your latest masterpiece into LM Studio and getting great speeds on my Nvidia Blackwell RTX PRO 2000 Laptop GPU.....running claude code locally! best model yet for my use case!!! THANK YOU! with Lenovo P1 gen 8!

[Hugo Bowne-Anderson @hugobowne]

What can we learn about parallelizing AI agent swarms from the Amish?

We had a wild 2.5 hour (Agent) skills sharing session yesterday with @hmason, @BEBischof, @ericmjl, &
@ttunguz.

A few highlights for me were

- How Agent can help bring science back into engineering (thanks, Hilary and Hamming!)
- How to build a grammar of graphics for agents (thanks, Bryan!)
- How to pair program with agents for exploratory data analysis in notebooks (thanks, Eric!)
- Building agent skills to analyze public companies using local models and the Pi agent harness (thanks, Tom!)

So much more that Thomas Wiecki, PhD (PyMC Labs and I are still processing. These are some of the things that were demod and discussed 👇

Full livestream here: https://t.co/myzU7p9eqE