HF-Party (Hugging Face Party @ PyTorch Conference)

posted an update about 1 month ago

Post

215

Released lafzyn , built over Qwen, an Urdu language model that converts Urdu text into IPA phonetic transcription, with GGUF builds for local inference.

Release contents:
- mahwizzzz/lafzyn: full weights
- mahwizzzz/lafzyn-gguf: quantized builds

Try it out 🤗
Demo: https://huggingface.co/spaces/mahwizzzz/Lafzyn

woojun-jung

authored a paper 2 months ago

QEVA: A Reference-Free Evaluation Metric for Narrative Video Summarization with Multimodal Question Answering

Paper • 2604.24052 • Published Apr 27

parkneurals

authored a paper 3 months ago

KV Cache Recycling to Expand Usable Context Capacity in Low Parameter LLMs

Paper • 2512.11851 • Published Dec 4, 2025 • 1

1aurent

authored a paper 3 months ago

Voxtral TTS

Paper • 2603.25551 • Published Mar 26 • 63

xianbao

submitted a paper to Daily Papers 4 months ago

The Curse and Blessing of Mean Bias in FP4-Quantized LLM Training

Paper • 2603.10444 • Published Mar 11 • 12

1aurent

authored a paper 5 months ago

Ministral 3

Paper • 2601.08584 • Published Jan 13 • 63

JingzeShi

authored 2 papers 5 months ago

Towards Automated Kernel Generation in the Era of LLMs

Paper • 2601.15727 • Published Jan 22 • 19

OmniMoE: An Efficient MoE by Orchestrating Atomic Experts at Scale

Paper • 2602.05711 • Published Feb 5 • 13

JingzeShi

submitted a paper to Daily Papers 5 months ago

OmniMoE: An Efficient MoE by Orchestrating Atomic Experts at Scale

Paper • 2602.05711 • Published Feb 5 • 13

julien-c

submitted a paper to Daily Papers 5 months ago

Shaping capabilities with token-level data filtering

Paper • 2601.21571 • Published Jan 29 • 29

ybelkada

authored a paper 6 months ago

Learnable Multipliers: Freeing the Scale of Language Model Matrix Layers

Paper • 2601.04890 • Published Jan 8 • 44

csabakecskemeti

posted an update 6 months ago

Post

3394

Just sharing a result of a homelab infrastructure experiment:

I've managed to setup a distributed inference infra at home using a DGX Spark (128GB unified gddr6) and a linux workstation with an RTX 6000 Pro (96GB gddr7) connected via 100Gbps RoCEv2. The model I've used (https://lnkd.in/gx6J7YuB) is about 140GB so could not fit either of the GPU. Full setup and tutorial soon on devquasar.com

Screen recording:
https://lnkd.in/gKM9H5GJ

3 replies

·

woojun-jung

authored a paper 7 months ago

Visual Funnel: Resolving Contextual Blindness in Multimodal Large Language Models

Paper • 2512.10362 • Published Dec 11, 2025 • 1

csabakecskemeti

posted an update 7 months ago

Post

1502

FYI: Mistral.Ministral-3 dequantizer FP8->BF16

https://github.com/csabakecskemeti/ministral-3_dequantizer_fp8-bf16

(The instruct model weights are in FP8)

csabakecskemeti

posted an update 7 months ago

Post

2112

Looking for some help to test an INT8 Deepseek 3.2:
SGLang supports Channel wise INT8 quants on CPUs with AMX instructions (Xeon 5 and above AFAIK)
https://lmsys.org/blog/2025-07-14-intel-xeon-optimization/

Currently uploading an INT8 version of Deepseek 3.2 Speciale:
DevQuasar/deepseek-ai.DeepSeek-V3.2-Speciale-Channel-INT8

I cannot test this I'm on AMD
"AssertionError: W8A8Int8LinearMethod on CPU requires that CPU has AMX support"
(I assumed it can fall back to some non optimized kernel but seems not)

If anyone with the required resources (Intel Xeon 5/6 + ~768-1TB ram) can help to test this that would be awesome.

If you have hints how to make this work on AMD Threadripper 7000 Pro series please guide me.

Thanks all!

8 replies

·

csabakecskemeti

posted an update 8 months ago

Post

319

Recently there are so much activity on token efficient formats, I've also build a package (inspired by toon).

Deep-TOON

My goal was to token efficiently handle json structures with complex embeddings.

So this is what I've built on the weekend. Feel free try:

https://pypi.org/project/deep-toon/0.1.0/

xianbao

authored a paper 8 months ago

RoboChallenge: Large-scale Real-robot Evaluation of Embodied Policies

Paper • 2510.17950 • Published Oct 20, 2025 • 9

csabakecskemeti

posted an update 9 months ago

Post

2668

Christmas came early this year

3 replies

·

mbrack

authored a paper 9 months ago

UniFusion: Vision-Language Model as Unified Encoder in Image Generation

Paper • 2510.12789 • Published Oct 14, 2025 • 19

jeffboudier

posted an update 10 months ago

Post

3446

Quick 30s demo of the new Hub > Azure AI integration to deploy HF models in your own Azure account. Now with Py and CLI!

GG @alvarobartt @kramp @pagezyhf

AI & ML interests

Team members 200

HF-Party's activity