Prithiv Sakthi's picture

Building on HF

Prithiv Sakthi PRO

prithivMLmods

hugging-science

·

https://linktr.ee/prithivsakthi

AI & ML interests

computer vision, nlp, multimodality - HuggingFace Fellow🤗

Recent Activity

updated a collection about 8 hours ago

Unredacted MAX - VL [FP8]

updated a collection about 8 hours ago

Unredacted MAX - VL [FP8]

updated a Space about 12 hours ago

prithivMLmods/Super-OCRs-Demo

View all activity

Organizations

upvoted a paper 1 day ago

VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training

Paper • 2602.10693 • Published 14 days ago • 167

upvoted 2 articles 1 day ago

Article

Kimi K2.5: Still Worth It After Two Weeks?

1 day ago

•

6

Article

I Let a Lobster Run My Jetson: What OpenClaw Taught Me About the Future of Computing

6 days ago

•

13

upvoted a collection 2 days ago

Qwen 2.5 VL Unredacted MAX [Feb 23, 26]

Continual “abliteration” models – experimental. • 6 items • Updated 1 day ago • 1

upvoted an article 4 days ago

Article

Forge: Scalable Agent RL Framework and Algorithm

12 days ago

•

126

upvoted 3 papers 4 days ago

SpargeAttention2: Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-Tuning

Paper • 2602.13515 • Published 11 days ago • 42

DDiT: Dynamic Patch Scheduling for Efficient Diffusion Transformers

Paper • 2602.16968 • Published 6 days ago • 11

Mobile-Agent-v3.5: Multi-platform Fundamental GUI Agents

Paper • 2602.16855 • Published 10 days ago • 42

upvoted a collection 5 days ago

Qwen Image Edit - AIOs / Compression

Collection of FP8 compression techniques, Diffusers–Transformers converted weights, and more for QIE. • 11 items • Updated 4 days ago • 1

upvoted 2 papers 6 days ago

Multi-agent cooperation through in-context co-player inference

Paper • 2602.16301 • Published 7 days ago • 22

RynnBrain: Open Embodied Foundation Models

Paper • 2602.14979 • Published 11 days ago • 40

upvoted 2 papers 8 days ago

CoPE-VideoLM: Codec Primitives For Efficient Video Language Models

Paper • 2602.13191 • Published 11 days ago • 29

OneVision-Encoder: Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence

Paper • 2602.08683 • Published 16 days ago • 48

upvoted a collection 9 days ago

Qwen3.5

6 items • Updated about 15 hours ago • 317

upvoted a paper 10 days ago

χ_{0}: Resource-Aware Robust Manipulation via Taming Distributional Inconsistencies

Paper • 2602.09021 • Published 15 days ago • 25

upvoted an article 10 days ago

Article

LateOn-Code & ColGrep: LightOn unveils state-of-the-art code retrieval models and code search tooling

13 days ago

•

46

upvoted 3 collections 10 days ago

Unredacted MAX - VL [FP8]

[FP8_DYNAMIC] multi-stage trained continual "abliteration" approach for the qwen3-vl series models • 10 items • Updated about 8 hours ago • 2

Unredacted MAX - VL

multi-stage trained continual "abliteration" approach for the qwen3-vl series models • 4 items • Updated 9 days ago • 4

Unredacted MAX - VL [GGUF]

multi-stage trained continual "abliteration" approach for the qwen3-vl series models (GGUF) • 4 items • Updated 10 days ago • 3

upvoted a paper 12 days ago

Towards Autonomous Mathematics Research

Paper • 2602.10177 • Published 14 days ago • 36