Ionut-Vlad Modoranu

PhD student @ ISTA 🇦🇹
Former Research Intern @ Together AI 🇳🇱
Studied Computer Engineering & Machine Learning 🇷🇴

[Dan Alistarh @DAlistarh]

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 @ionutmodo]

@OncelTuzel Hi, @OncelTuzel! Thank you for posting this announcement! I have just applied to this job as it fits my PhD expertise.

[Ionut-Vlad Modoranu @ionutmodo]

@calcsam BOOK

[Ionut-Vlad Modoranu @ionutmodo]

@asmah2107 Hi! I'm currently a PhD student working in optimizers for deep learning. I'm focused on decreasing the memory usage and speeding up the runtime.

[Ionut-Vlad Modoranu @ionutmodo]

@iamgrigorev This is a great tip for productivity! Please check out our GridSearcher project, it is completely written in Python and allows you to run jobs in parallel or sequentially using dictionaries for hyper-parameters by employing a basic scheduling: https://t.co/t4dUUuxs5g

[Dan Alistarh @DAlistarh]

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: https://t.co/ulbX5D5LjD
Code: https://t.co/OYfjs4jdDp

[Dan Alistarh @DAlistarh]

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]

[Dan Alistarh @DAlistarh]

🚀 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 @ionutmodo]

@KwangjunA This seems to be similar to our recently shared work, where we use Discrete Cosine Transform (DCT) to perform a cheap low-rank projection of the momentum buffer, followed by Newton-Schulz orthogonalization. Check out our paper here: https://t.co/Sao5sBDDTu

[Dan Alistarh @DAlistarh]

Paper: https://t.co/9jwxD1tU92
We release vLLM & HuggingFace integrations.
Code: https://t.co/pvkXTyDlcO
Kernels: https://t.co/KjO04HAHyq

Credit goes to the team: @RobertoL_Castro, Vage, Denis, @black_samorez and @AshkboosSaleh, with support from @RedHat_AI and @thoefler

[Ionut-Vlad Modoranu @ionutmodo]

@OptionsBuffett Trade

[Ionut-Vlad Modoranu @ionutmodo]

@deedydas DeepSeek gets it right:

[Dan Alistarh @DAlistarh]

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:
https://t.co/JjyYT9Mrun