vitoplantamura/OnnxStream
View on GitHub ↗ ★ 2,087 C++
Lightweight inference library for ONNX files, written in C++. It can run Stable Diffusion XL 1.0 on a RPI Zero 2 (or in 298MB of RAM) but also Mistral 7B on desktops and servers. ARM, x86, WASM, RISC-V supported. Accelerated by XNNPACK. Python, C# and JS(WASM) bindings available.
7 mentions on Hacker News by 4 people: 5 stories, 2 comments · first 2023-07-20 · latest 2026-02-16
20082026
-
We were looking into OnnxStream ( https://github.com/vitoplantamura/OnnxStream ) and modifying it to support img2img. We got pretty close but yeah capability of running diffusion models on a Raspi are quite limited lol. Alternatively we…
comment by alexkranias · 2026-02-16 · in Show HN: Diffuji – a diffusion-powered instant camera -
▲ 2
Show HN: OnnxStream running TinyLlama and Mistral 7B, with CUDA supportstory by vitoplantamura · 2024-01-14 · 0 comments
-
▲ 17
OnnxStream running TinyLlama and Mistral 7B, with CUDA supportstory by Robin89 · 2024-01-14 · 2 comments
-
…be correct! I took "Stable Diffusion XL 1.0 Base on a Raspberry Pi Zero 2" from the About part of https://github.com/vitoplantamura/OnnxStream/tree/c0cb4b3d7b... . Is it wrong?
comment by dang · 2023-12-14 · in OnnxStream: Stable Diffusion XL 1.0 Base on a Raspberry Pi Zero 2 -
▲ 109
OnnxStream: Stable Diffusion XL 1.0 Base on a Raspberry Pi Zero 2story by Robin89 · 2023-12-14 · 87 comments
-
▲ 510
Running Stable Diffusion XL 1.0 in 298MB of RAMstory by Robin89 · 2023-10-03 · 154 comments
-
▲ 293
Running Stable Diffusion in 260MB of RAMstory by Robin89 · 2023-07-20 · 64 comments