https://xitter.cf/t.co/9pARQ3GP0r • SVP Technology & Architecture @ltnglobal • Previously @maketv, @defrac_ and @audiotool • Probably not the @joa you mentioned in your tweet
Bielefeld, Germany
Joined September 2007
- Tweets4.7K
- Following544
- Followers3.6K
- Likes648
phobos-lang.dev runs @OpenBMB MiniCPM MoE Q8_0 pp128 at 6x of llama.cpp 😅
Several compiler optimizations landed in the past two weeks and inference certainly benefits. Also faster on 3090, 4090 or 5090 cards now.
And by the way 🎤🫳 : phobos-lang.dev/inference/
Phobos has now a home that's a bit more polished than a bare github repo :)
Æon learned to talk. My purely functional audio engine now has a formant speech synth and a vocoder. No samples. Just a (very basic) model of a throat as a function of t.
Then I asked Opus 5.5 for a track. It wrote the lyrics as phonemes on a beat grid:
S IH S | T AH M | AA N | L AA IY N
That's "system online".
It also included a bit of uncanny existential crisis: "are we machines, or are we dreaming?". You can listen to machine_heart.js, a purely functional song, at joa.github.io/aion
My GPU is from 2019 and has 8 GB of VRAM. It now runs a 35B MoE that doesn't fit on it at 55 tok/s.
Every kernel is compiled at runtime from Phobos, a toy kernel language I wrote to learn what a kernel even is. No cuBLAS, no CUDA toolkit. Just the driver.
No knobs. Phobos sizes its expert cache by free VRAM and the CPU computes the misses. When the desktop grabs 2 GiB mid-session, it keeps ~75% of its decode speed. llama.cpp at its fastest setting keeps 30% (-ncmoe 36 survives it, but runs slower all the time).
Is it better than llama.cpp? Absolutely not. llama.cpp runs everywhere, and I optimized for one card hardly anyone cares about. But for that card I can tell you exactly where the time goes.
Blog Post: joa-ebert.com/posts/2026-10-…
v0.1.1, Windows and Linux: github.com/joa/phobos-lang
This is @typesafeai #jev on the decks. I clearly need to tweak the context more. All parameters / decisions are driven by the model. It's similar to Jev playing Doom with a decision tree of what to do.
I added @OmarchyLinux mode to Phobos. Because it's that's what you do in 2026 🤘(I had to use a tiny model, as otherwise the vid capture wouldn't work 😅)
Use Jev to reduce the comment vomit in your code. 🤢 github.com/joa/code-compact-…
Can anyone recommend a cheap platform that allows me to rent a H100 where I can run ncu and nsys for perf tuning?
