William Smith

@claudeai My wife is in special Ed and is a director for an Autism clinic. She is a current credentialed educator! Can she apply? ALSO she uses Claude Code like a champ every single day!

We're introducing Claude for Teachers: free access to premium Claude capabilities for verified K-12 educators in the US, with a library of teaching skills and a direct connection to evidence-based curricula, mapped to academic standards in all 50 states. claude.com/solutions/teac…

@MiaAI_lab @NVIDIAAI GLN is a pretty damn good model! It’s my fallback for my main hive agent! I fool a lot of tool calling and it’s one of the best! I’ve switched to Hermes and it’s really solid! I wouldn’t hesitate to use it for any of my Enterprise Clients!

What are the best models you can run on your @NVIDIAAI DGX Spark? ✨ Mid-July 2026 Edition 1× DGX Spark • ⁠Qwen 3.6 35b NVFP4 — 256k ctx, 81 tok/s • ⁠Qwen 3.6 27b NVFP4 — 256k ctx, 33 tok/s 2× DGX Sparks ← sweet spot! • DeepSeek v4 Flash — 1M ctx, 60 tok/s • MiMo-V2.5 — 1M ctx, Full Omni, 31 tok/s • Step-3.7-Flash — 256K ctx, 30 tok/s 3× DGX Sparks With three units, I recommend running a large model across two of them and a smaller one on the third. I'd personally run DeepSeek v4 Flash + Qwen 3.6 35B. This setup gives you DeepSeek v4 Flash speeds for coding, plus strong agentic workflows and image support from Qwen 3.6 35B. 4× DGX Sparks GLM 5.2 NVFP4 across all 4 units — if you have four units, this is the one to run! Links and repos below 👇

@mr_r0b0t @llmrequirements @vllm_project @NVIDIAAI What are you preparing it to be used in?

@mr_r0b0t @vllm_project @NVIDIAAI How is it at prompt adhearence, long train of thought? Batch processing? Concurrency? Have you used it in an actual production workflow?

SM121 optimized @vllm_project v0.25.0 producing a noticeable increase in decode speeds when serving @NVIDIAAI-Qwen3.6-35B-A3B-NVFP4! 93.05 tok/s on a single DGX Spark (GB10) at MTP=2 full engine configuration and results: github.com/r0b0tlab/qwen3…

@Gaurav_dev01 And that’s where your latency will come from.

@taleosmith Well said. The model is just one component. The real product is everything built around it. 👏

Everyone is chasing smarter AI. I'm more interested in making AI boring. Reliable. Predictable. Cheap. That's what companies actually pay for.

@yasfdany Followed!

Just got laid-off today... I’m open to any Software Engineer roles, especially in Mobile Development (Flutter/React Native).

Let’s help one of us out!!!!

@brockpierson My DGX Spark

What's one purchase you've never regretted?

@Nishanttt05 Copilot

Be honest Which AI model is worthless to use in 2026?

@jacoblabsai Gotta pay attention to that non-compete and NDA!!

Hiring competitor talent is normal. Being unable to explain where the new work came from is the real problem.

@markproduct Wow, about 2-3am most nights!

What almost made you quit building forever?

@emilstridell @loktar00 Just posted the same!

@loktar00 3090 is the same price with way more perf. those are only 8gb each.

3x3060ti mining rig for sale for $1000.... am I going insane for considering this?! Asking the seller how much ram/storage, it's a mining board and connected via x1's it looks like but 1k for the gpu's alone isn't a bad deal.

@loktar00 Beware! Go on eBay and grab a 3090! Boom done!

@Blackwellboy Ordered mine and didn’t have any issues. Good to know though!

First time I order from the USA. My ASUS GX10 turn up almost a week early. Great! No. Not great. Seeing a few peoples tweets about issues receiving these from Amazon damaged, I decided to record from grabbing it off of my porch to the unboxing. I suspected bad packaging, possibly damaged, but I didn’t suspect no unit inside. When I initially saw it I noticed some sides of the box the tape was up and the box itself looked like it had been through ww3. I picked it up, it felt a bit light but heavy enough not for my heart to sink too much. Anyway long story short, everyone that knows me knows I keep my packaging pristine. I open the box pull it out, seal was broken, box looked like this and no unit. Charger still inside though guess that’s why it felt a bit heavy. Now amazon asking for a police report and all that bs. Anyone know of anyone with bad experiences like this or how the process works please reach out. What a day.

@stuckAgainDev Outage

I am a claude user, scare me with one word😏

@13_narcissus I’d rather not say!

How many AI subscriptions are you paying for right now? I'll start: 2 (claude code + codex)

@PaulGugAI Kinda seems like that’s all people do on x is benchmark! No one says if they actually use them in workflows or agents!

Benchmarks are not everything, but still an important criteria that needs to be met. This is very positive in terms of healthy competition, and bodes well for the likes of Grok 5. Can't wait to see what a super-large Grok LLM looks like.

@mr_r0b0t What are you going to use it for?

Pleased to say this Hy3 NVFP4 is looking quite performant 🤩

@lockedindev_ Trying

Are you gaining new followers or just scrolling

@i_mika_el Nope 2 million sets me free from all other humans! And can build my dream home!

would you rather sell your product for $2M tomorrow or turn it down and bet on it becoming $50M