ONNX
Fewer than 5 people recommend it so far, so there's no written verdict. Here is what commenters said.
3recommend
2warn against
Dec 25, 2025first mention
Oct 7, 2026latest mention
Why people recommend it
“We use the ONNX runtime for small models in the browser”
“We use this in production:”
serves 5+ million inference requests per dayon “OpenCV 5 Is Here: The Biggest Leap in Years for Comp…”
Show 1 more
“onnx on the web has the most models available and can use webgpu which is available everywhere.”
Caveats and dissent
“dealing with ONNX is kind of annoying”
built a cross-platform live captioning appon “Transcribe.cpp”
“My experiences with ONNX have not been pleasant. Conversions from models written with Tensorflow and Pytorch often fail.”
Who recommends it in their own words
Threads where it came up
- Jev-Driven SRE Diagnosis: What Worked and What Failed 1 mention
- WebLLM: high-performance in-browser LLM inference engine 1 mention
- Show HN: NegativeEV – I built a tool to help people see how bad their bets are 1 mention
- Transcribe.cpp 1 mention
- Moebius: 0.2B image inpainting model with 10B-level performance 1 mention
- OpenCV 5 Is Here: The Biggest Leap in Years for Computer Vision 1 mention
Also written as ONNX Runtime. Data: onnx.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 8
“onnx scores all my HA entities to determine which one I'm talking about”
“We use the ONNX runtime for small models in the browser”
“it's so the Rust simulator can run the model itself, with no Python in the loop”
“dealing with ONNX is kind of annoying”
“I got this working with ONNX”
“We use this in production:”
“onnx on the web has the most models available and can use webgpu which is available everywhere.”
“My experiences with ONNX have not been pleasant. Conversions from models written with Tensorflow and Pytorch often fail.”