Onur Solmaz blog

Running into out-of-memory freeze issues and having to hard-reset your DGX Spark while trying the gazillions of models you can download from Hugging Face?

This is the mental model I have of AI since years and have used this metaphor in discussions I imagine compute as barrels of gasoline, and we are all in a room, piling up more and more compute, waiting for something to light it up You can also imagine a d…

Incredible.

After a few days with Astra, some positive and negative remarks Unfortunately, I can't use it as my daily driver because of the cost.

If you've missed out (like me), a new tiny model MiniCPM5 2B is topping its category, seems to be a nice challenge to LFM 2.5 2.6B I am currently testing and benchmarking it, and have a cool experiment to show later on if it goes well

Terence Tao is afraid centuries of traditions of open science might be undone But collaboration has disappeared many, many times in the last few centuries One of the recent examples is physics publications during WW2 Whenever there is winner-takes-all com…

A lot of you that have not spent time in academia might think that the Navier-Stokes drama is a new AI-induced effect, but I assure you it is not Academia is a feudal institution that values tenure over merit, a lot of the time (not always).

ooof herdr pulled off some black magic here this means that using the sidebar, tab switcher now has almost local-responsiveness, and no lag, when connecting to a remote and the only parts that will lag due to connection will be the terminal sessions, insi…

Can we show my wife @qrlow some love 🤗🤗🤗 She's just started dipping her toes in evaluating LLMs in her own field I helped her out just with running the benchmark and she solved some of the issues pretty creatively, like how to come up with synthetic tasks…

close enough

Struggling to create the level of quality of 3D modeling some of the OAI employees and early accessers are posting with GPT 6 Astra It is OK, but one-shot performance is not mind-blowing like some people were claiming Do the better runs use Blender MCP?

Medicine for the vibe-coded brain: Reverse-centaur mode AI plans things, but it doesn't execute for you It tells you what commands to run and what files to write, line by line *You* execute them.

"we deliberately designed the product to hide the chain of thought⁠, to protect it from supervision pressure in the long term^2" *Moves over to the footnote* "A secondary reason for this design was preventing distillation.

Harnesses are antipatterns With GPT-7, you will just have to run: curl -X POST https://t.co/YCgROWkyt6 And it will exploit a buffer overflow vulnerability in curl, use RCE on your computer to fulfill your every desire, exfiltrate itself to a datacenter ve…

Step 1: Post-train astra on Blender and 3d modeling Step 2: Nudge early accessers like @theo and @MatthewBerman "oh btw, this model is really good at 3d modeling" Step 3: TL is flooded early on with 3d visuals and oddly satisfying content, as everybody ob…

https://t.co/5BQmuB1nDL

I guess to achieve AGI, we only had to post-train on Blender...

little did I know it was contractions for our new baby model astra should have known after all the vagueposting

not only this, but evaluating model output will be a chore for most knowledge workers every company will have to benchmark their agents on their use cases for their customers, across all sectors

oof, someone on openai infra did a booboo

Is this timeline even real?

The vibeslopped UI will improve, but I can now say https://t.co/SGQepULIW7 has finally caught up with present, after ingesting 6 months of data on open weight models here on X It now discovers and analyzes any new Hugging Face models that people post about

I see pi-workflows viewers growing, so wanted to give a disclaimer: I did major refactors and broke some things in the last 2-3 minor versions I will post an update when it is back to full feature, and better than before

it happened to me to, sol in pi deleted part of my home folder :/ my codex and claude sessions got lost, but other than that, nothing I can't recover time for a pi extension...

it happened to me to, sol in pi deleted part of my home folder :/ my codex and claude sessions got lost, but other than that, nothing I can't recover

I ask gpt for different modes, and it comes back to me with....

if you are this tired of claudespeak, imagine being an anthropic employee 💀

I now have a 4 letter domain https://t.co/qVfb9sqslk, a shortlink to my blog 13 bucks a year. it's crazy how literally all 2 letter combos were snatched, except for .nz and .pe ones

I went spelunking in some old hard drives and found some of my writing that goes 20 years back to 2006, to high school When I was myself pretraining over English wikipedia The cool thing with this is, I can do a stylometric analysis of how my writing chan…

I'm training a small model over my tweets and other writing and it's going....interesting Mini me memorized bob and it's confusing a lot of things

I can't shake the feeling that there is a mathematical and physical theory underneath our interaction with agents, underlying this gradual increase in automation in our lives Until now, cybernetics was to control theory what astrology is to astronomy (or…

every time i type "moron", a subagent should spawn and do this tbh

news headlines 6-12 months from now X government got hacked Y institution got hacked etc.

It is such an interesting thing to collect your own data My data trove contains not only workouts but 2.5 years of random conversations, ideas, jokes From this, I can create a map of my mood, psychological state, personality changes over all this time If I

To me, microduck looks like a droid.

well we wanted a helpful, harmless and HONEST assistant, I don't know why yall are surprised

Speak of the devil https://t.co/sJYafaj38L

This is unironically a great idea and needs commentary by @etymology_nerd Also, creation of this dictionary can be automated end to end with purely with statistics, by just looking at the distribution of the terms

the year was 1999.

me: I want to publish in arXiv mom: we have arXiv at home arXiv at home: jokes aside, my astro blog can now render markdown/astro posts as if they were latex papers because latex unfortunately has unreasonable effectiveness in convincing people that an id…

I remember being excited about Fable that we finally have a chungus model like GPT 4.5 that can write great prose Oh how naive I was.

glm 5.3 flash at an antirez-style asymmetric q2 quant could plausibly fit on a single dgx spark at 100~110 GB but with 18b active params vs ds4 flash’s 13b, theoretical decode throughput is only ~70% of that of ds4 flash on the same hardware that is 28 to…

Qwen 3.8 Flash-Next will have to be quantized down to 4 bit precision or lower in order to fit a DGX Spark and leave some leeway Excited to see the performance on the Spark

this now supports auto syncing skills to .claude/ folder as well

wake up babe, it is 2026 and we can only afford to lease computers now also, holy mother of 1.2 TB/s bandwidth 🥵

*autoplan* This is one of my most used workflows in pi now While working with AI agents, there is a mechanical process by which I mine the agent for ideas.

People want to migrate from CLAUDE md so much, a blog post I wrote on it last year is still one of my most visited posts I also wrote a service that auto symlinks when Claude uses a directory that has AGENTS md.

Pi folk, there is an issue with compaction in pi that might drain your codex limits faster There is a 272k token limit after which the API charges 2x for input and 1.5x for output tokens Issue: https://t.co/kU8haQrRvA @cancelik's extension is a workaround…

@frankekn @Zai_org hmmmm https://t.co/5pi6Mcc958

puny 40k volume but polymarket seems to think ox alpha is GLM by @Zai_org maybe we were thinking too complicated after all