cTuning & MLCommons Collective Knowledge Challenges

[ Back ]

---

Run and optimize MLPerf inference v2.1 benchmarks

Open date: 2022 Jul 1

Closing date: 2022 Sep 1

Shared experiments:

• [mlperf-inference,v2.1](?action=experiments&tags=mlperf-inference%2Cv2.1)

---

Challenge

Prepare, optimize and reproduce MLPerf inference v2.1 benchmarks across diverse implementations, software and hardware using the MLCommons CK framework.

Organizers

• MLCommons taskforce on automation and reproducibility
• cTuning foundation
• OctoML

Status

This challenge has been successfully completed.

Results

Results are available here.

---

Self link