PhD student @ ISTA 🇦🇹 Former Research Intern @ Together AI 🇳🇱 Studied Computer Engineering & Machine Learning 🇷🇴
Vienna
Joined April 2018
- Tweets16
- Following290
- Followers57
- Likes98
Training LLMs just got faster
I'm at @icmlconf in Seoul this week presenting DASH - our latest improvement to the well-known Shampoo optimizer. By using DASH, we're achieving 5x lower running time and higher accuracy than the standard Distributed Shampoo implementation!
[1/3]
Ionut-Vlad Modoranu retweeted
We're releasing DASH (Distributed Accelerated Shampoo), an improved implementation of the Shampoo optimizer that achieves up to 4.83× faster optimizer steps, while matching or improving final model quality.
[1/6]
Ionut-Vlad Modoranu retweeted
Our QuEST paper was selected for Oral Presentation at ICLR @sparseLLMs workshop!
QuEST is the first algorithm with Pareto-optimal LLM training for 4bit weights/activations, and can even train accurate 1-bit LLMs.
Paper: arxiv.org/abs/2502.05003
Code: github.com/IST-DASLab/QuEST
Ionut-Vlad Modoranu retweeted
We're releasing the DASLab GGUF Quantization Toolkit! 🚀
First open-source toolkit bringing GPTQ + EvoPress to @ggerganov's GGUF format, enabling heterogeneous quantization based on importance.
Result: Better models at the same file size.
[1/5]
Ionut-Vlad Modoranu retweeted
🚀 We are releasing state-of-the-art post-training quantization (PTQ) algorithms for Microscaling FP4, together with kernels:
- First study focused on MXFP4/NVFP4 PTQ for LLMs
- New Micro-Rotated (MR) format and GPTQ algorithm
- QuTLASS GPU kernels with up to 3.6x speedups.
Ionut-Vlad Modoranu retweeted
Paper: arxiv.org/abs/2509.23202
We release vLLM & HuggingFace integrations.
Code: github.com/IST-DASLab/FP-Qua…
Kernels: github.com/IST-DASLab/qutlas…
Credit goes to the team: @RobertoL_Castro, Vage, Denis, @black_samorez and @AshkboosSaleh, with support from @RedHat_AI and @thoefler
Ionut-Vlad Modoranu retweeted
Introducing Panza, a personalized LLM email assistant, running entirely on-device!
[1/6]
* Panza adapts LLaMA-3-8B to match your unique writing style;
* Can be fine-tuned and executed on a single GPU (free Colab version available!).
Give it a try:
github.com/IST-DASLab/PanzaM…
