⚠️ This repo is stale — the MyPCBench VM image lives in a dataset repo

The daily-rebuilt, eval-canonical MyPCBench VM image (michael_scott.qcow2) is published here:

➡️ https://huggingface.co/datasets/ljang0/mypcbench-qemu-baseline

That dataset repo is rebuilt and republished daily and is byte-identical to the qcow2 embedded in the Docker image ljang/mypcbench-qemu:latest (verified by a freshness check). The benchmark task set + rubrics are at https://huggingface.co/datasets/ljang0/MyPCBench-tasks.

This same-named model repo is an older orphaned copy kept only so existing links don't 404. Do not use it — fetch the image from the dataset repo above.

Code: https://github.com/ljang0/MyPCBench
Paper: https://huggingface.co/papers/2606.16748

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Paper for ljang0/mypcbench-qemu-baseline

MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents

Paper • 2606.16748 • Published 25 days ago • 7