Zach Mustafa's picture

Zach Mustafa PRO

Zmu

·

AI & ML interests

None yet

Recent Activity

liked a model 9 days ago

Qwen/Qwen3.5-0.8B-Base

liked a model 9 days ago

Qwen/Qwen3.5-2B-Base

liked a Space 13 days ago

davanstrien/dataclaw-conversations-atlas

View all activity

Organizations

upvoted an article 13 days ago

Article

GGML and llama.cpp join HF to ensure the long-term progress of Local AI

+4

20 days ago

•

480

upvoted a collection 21 days ago

Qwen3.5

21 items • Updated 2 days ago • 1.12k

upvoted an article 26 days ago

Article

NVIDIA brings agents to life with DGX Spark and Reachy Mini

+1

Jan 5

•

65

upvoted 3 collections about 1 month ago

Heretic - Abliterated, Uncensored, Unrestricted POWER.

Models that have be abliterated using the HERETIC method. Done properly, this completely removed almost all censorship with no damage to the model. • 91 items • Updated about 1 hour ago • 40

✂️ Abliteration

Uncensored models using abliteration. See this article for more information: huggingface.co/blog/mlabonne/abliteration • 32 items • Updated 10 days ago • 162

Qwen3-VL-abliterated

14 items • Updated 10 days ago • 31

upvoted an article about 1 month ago

Article

Introducing Daggr: Chain apps programmatically, inspect visually

+3

Jan 29

•

103

upvoted 3 collections about 2 months ago

VibeVoice

Frontier Text-to-Speech Models https://microsoft.github.io/VibeVoice/ • 8 items • Updated 10 days ago • 211

Qwen3-TTS

7 items • Updated Jan 22 • 323

DroPE

Extending the Context of Pretrained LLMs by Dropping Their Positional Embedding (https://www.arxiv.org/abs/2512.12167) • 1 item • Updated Jan 11 • 2

upvoted a collection 3 months ago

Qwen3-Omni

6 items • Updated Dec 31, 2025 • 187

upvoted 3 articles 3 months ago

Article

New in llama.cpp: Model Management

Dec 11, 2025

•

129

Article

Transformers v5: Simple model definitions powering the AI ecosystem

+2

Dec 1, 2025

•

305

Article

Continuous batching from first principles

+1

Nov 25, 2025

•

343

upvoted 3 collections 4 months ago

Olmo 3

Artifacts for the Olmo 3 release. • 7 items • Updated 10 days ago • 166

V-JEPA 2

A frontier video understanding model developed by FAIR, Meta, which extends the pretraining objectives of https://ai.meta.com/blog/v-jepa-yann • 8 items • Updated Jun 13, 2025 • 195

ViDoRe Benchmark V3

ViDoRe V3 is our latest benchmark, engineered to set a new industry gold standard for multi-modal, enterprise document retrieval evaluation. • 8 items • Updated Jan 14 • 20

upvoted a paper 6 months ago

zELO: ELO-inspired Training Method for Rerankers and Embedding Models

Paper • 2509.12541 • Published Sep 16, 2025 • 9

upvoted a collection 6 months ago

OpenVision

27 items • Updated Aug 15, 2025 • 33

upvoted a collection 7 months ago

DINOv3

DINOv3: foundation models producing excellent dense features, outperforming SotA w/o fine-tuning - https://arxiv.org/abs/2508.10104 • 15 items • Updated 1 day ago • 533