Suryaa retweeted
Appreciate it @MonacoGTM
3rd at the invitational next time I’m coming to win all my money back
Connected to kimi, Ff5-tts, wan2.2. Ran models with ionrouter.io only using one image and 8 seconds of my voice. only costed me $0.20in/$1.60out for Kimi on it. Can share repo if ppl interested 🔥
2
156
Suryaa retweeted
We just launched IonRouter: ionrouter.io
One API for the best open‑source models — LLMs, vision models, TTS, and video generation.
Suryaa retweeted
they say cambridge is the smartest sq. mi in the world but what about impactful? @CumulusLabsIO @ycombinator
Suryaa retweeted
Our intern just took 2nd place at the YC Hackathon with an AI editor that auto-edits inside DaVinci Resolve.
Ppl when they realize the hackathon idea they spent 5 hrs on already been made @ycombinator
1
5
183
We built on the Grace Hopper architecture from the ground up for our inference engine: ionAttention.
cumulus.blog/ionattention
Suryaa retweeted
Modal makes you wait. Together makes you pay for always-on.
Cumulus gives you both — serverless GPUs + a real inference stack (ION) that outperforms stock vLLM/SGLang with zero setup.
Pay by the second. Scale to 0. Cold starts under 10s.
Let's talk → calendar.app.google/DfDdq1fN…


















