Inference engine shrunk to 1023 bytes, still runs SmolLM2-135M
cneuralnetwork · x · 2026-11-04
utsavkhairnar previously squeezed an inference engine down to 2.9 KB—smaller than a C hello world (67,608 bytes) and small enough to fit in a QR code—running a real SmolLM2-135M model with no Python, no libraries, not even libc. A day later, cneuralnetwork pushed it under 1 KB (1,023 bytes), with a screenshot showing the engine completing a Python-style Fibonacci function. A PR is open.
More from Infra
- Self-improving software is inevitable: closing the agent-observability loop — bibryam · 2026-11-04
- Gigabyte's AI TOP motherboard packs 2TB RAM and 9 GPUs for home AI labs — vincent_koc · 2026-11-04
- Building a 7-GPU 132GB VRAM local inference rig: consolidate on EPYC or split across two systems? — infectiousstupidity · 2026-11-04
- imec neural probe integrates photonics directly into 130nm CMOS backend — jwt0625 · 2026-11-04
- Gary Marcus: AI data center demand has already passed its peak as projects get killed — GaryMarcus · 2026-11-04
- DeepMem: byte-level multiscale memory cuts GPU memory 54% with lossless decoding — Afshin Khadangi · 2026-11-04