this is where interactive media is heading
meet interview couch bear : he remembers what you told him last time. hears emotion in your voice. explains ideas with the objects around him.
all in real-time, powered by world models
made with @reactorworld
I built an inference engine that guesses ahead, checks its own work, and never changes the answer.
Introducing fast-rs: exact speculative decoding for the Mac you already own.
> 2–2.5× faster on full Spec-Bench
> 1,920/1,920 answers bit-identical
> 13″ M2 MacBook Pro, 8 GB 🧵
Introducing aq: framework for applied ML and foundation models
You can build, train, and check your models from one simple CLI or a few lines of Python. It covers everything from tabular models to transformers and LLM fine-tuning with LoRA.
Train, evaluate, track, and serve with a handful of commands. If you already have a training script, add a few lines of the aquin SDK, and your metrics and runs are tracked without rewriting anything.
When you need more power, connect your own machines over SSH, send jobs to them, and pull the results back, all from your laptop.
aq is open source. Install it in one line and start building.
💻 Code: github.com/aquinf03/aq
📄 Docs: aquinf03.github.io/aq/docume…
What if you could make games with AI?
we're hosting "MAKE IT PLAYABLE", a game creation challenge with @ElevenLabs, part of @chromaawards in Delhi NCR on 14th november.
Build games. Experiment. Make them playable. 🎮
sign up now: luma.com/cnlehsw4
what happens when the viewer decides what happens next in the story ?
we're hosting a kickoff talk for a film residency in USC on 9th nov (8 pm PT)
check out : luma.com/g5ozassp
Interactive storytelling is one of the most exciting spaces in media rn.
Excited to bring this conversation to USC and kick off BRANCH.
If you’re interested in the future of storytelling, come join us!
What if your audience could shape the story?
We’re kicking off BRANCH- a 2 week film residency at USC with @CtrlMovieLtd to explore interactive storytelling.
📅 Nov 9 | 8–9 PM
Register: luma.com/g5ozassp
We fine-tuned a talking-avatar LoRA on MiniMax-H3 Ref2VA: one character image plus dialogue becomes a clip with model-generated speech, lip sync, and light motion.
On ~385 clips with just 10.6M trainable parameters using aq. It handles portraits, paintings, and anime characters!
Model: huggingface.co/Aquinlabs/h3-…
Code: github.com/Aquinf03/h3-chara…
Launching MoM: Mixture of Models.
Large multimodal models are powerful but expensive to run. MoM is a cost-efficient alternative: compose smaller, specialized models and tools into one pipeline, and call the whole thing like a single model.
Build AI layers by combining different edge models (vision, speech, SLMs, decision, classifiers etc.) with Shared state, speculative routing and a ready-made API come built in, and the latency-critical path runs on Rust!
Code: github.com/Aquinf03/MoM
H3 Avatar aq is live on Hugging Face.
It takes a single reference image + dialogue and generates a speaking avatar video with identity preservation, expressive mouth motion, subtle head movement, and model-generated audio.
Try it here: huggingface.co/Aquinlabs/h3-…
trained on ~385 clips with just 10.6M trainable parameters.
It can animate everything from portraits and paintings to anime characters, with generated speech + lip sync.
Single image in → talking video out
ash is 17 and has already started over five times.
for him, the biggest support is his mom who has said yes more times than we can think of.
now he's doing interpretability research and running @AquinF03 labs
he was the youngest in our last cohort and has been hitting miles. so proud of this one!