@algoperf

AlgoPerf benchmark for faster neural network training via better training algorithms. Also on https://xitter.cf/t.co/ElK9Yeal7s

github.com/mlcommons/algorithmic-efficiency
Joined March 2025
  • Tweets16
  • Following23
  • Followers90
6Media
We just released AlgoPerf v1.0.1! 📚 New workload: AlgoPerf now includes a decoder-only 150M Language Model. ⚙️ Hardware switch: moving from 8xV100 to 4xA100. 🐞 Bug fixes and improved API. Coming soon: updated leaderboard with new optimizers! 👀
1
8
2
15
3,866
Learn more about AlgoPerf and this new release: github.com/mlcommons/algorit…

GitHub - mlcommons/algorithmic-efficiency: MLCommons Algorithmic Efficiency is a benchmark and...

MLCommons Algorithmic Efficiency is a benchmark and competition measuring neural network training speedups due to algorithmic improvements in both training algorithms and models. - mlcommons/algori...

github.com
4
183
We just released AlgoPerf v0.6! 🎉 ✅ Rolling leaderboard ✅ Lower compute costs ✅ JAX jit migration ✅ Bug fixes & flexible API Coming soon: More contemporary baselines + an LM workload… github.com/mlcommons/algorit…

GitHub - mlcommons/algorithmic-efficiency: MLCommons Algorithmic Efficiency is a benchmark and...

MLCommons Algorithmic Efficiency is a benchmark and competition measuring neural network training speedups due to algorithmic improvements in both training algorithms and models. - mlcommons/algori...

github.com
1
9
3
43
5,412
We're all about acceleration! 😉 Watch @KasimbegPriya & @frankstefansch1 speedrun an explanation of the AlgoPerf benchmark, rules, and results all within a tight 5 minutes for our #ICLR2025 paper video on "Accelerating Neural Network Training". See you in Singapore!
1
4
1
14
2,636
The explainer video: youtube.com/watch?v=_yX1ItxR…

ICLR 2025: Accelerating Neural Network Training (AlgoPerf)

-------------------------------------------------------------------...

youtube.com
1
293
AlgoPerf retweeted
AlgoPerf@algoperf17 Mar 2025
Replying to @reakyleru
Yes! We're switching to a rolling leaderboard, updated continuously without submission deadlines! We're currently scoring a tweaked baseline & past winners on an updated benchmark version (rule fixes, new runtime budgets, ...). Leaderboard updates will be posted on this account.
1
5
151
Hi there! This account will post about the AlgoPerf benchmark and leaderboard updates for faster neural network training via better training algorithms. But let's start with what AlgoPerf is, what we have done so far, and how you can train neural nets ~30% faster.
1
4
1
17
1,225
more replies
This is just the beginning! With the help of the community, we want to test even more training recipes, push the SOTA for neural network training methods even further, and improve the AlgoPerf benchmark. Follow us & join the working group (mlcommons.org/working-groups…).

Algorithms - MLCommons

The Algorithms working group creates a set of rigorous and relevant benchmarks to measure neural network training speedups due to algorithmic improvements.

mlcommons.org
1
3
153
More details: 📄 ICLR 2025 results paper: openreview.net/pdf?id=CtM5xj… 📄 Benchmark paper: arxiv.org/abs/2306.07179 💻 AlgoPerf codebase: github.com/mlcommons/algorit…
3
127
Load more