Valeriy Selitskiy's picture

Valeriy Selitskiy PRO

WaveCut

·

https://selit.ski

AI & ML interests

Looking to switch from hobby to career

Recent Activity

updated a model about 11 hours ago

WaveCut/FLUX.2-TE-Trimmed-7L-Distil

published a dataset about 17 hours ago

WaveCut/FLUX2-TE-Teacher-Embeds

published a model 1 day ago

WaveCut/FLUX.2-TE-Trimmed-7L-Distil

View all activity

Organizations

upvoted 2 articles 4 days ago

Article

Continuous batching from first principles

5 days ago

•

196

Article

Diffusers welcomes FLUX-2

5 days ago

•

135

upvoted 5 collections 9 days ago

KVAE 1.0

KVAE 1.0 tokenizers are for images (KVAE-2D-1.0) and video (KVAE-3D-1.0) are distributed under MIT license (commercial use is possible). • 2 items • Updated 5 days ago • 5

Kandinsky 5.0 Image Lite Diffusers

4 items • Updated 5 days ago • 8

GigaChat Lite

8 items • Updated 10 days ago • 2

GigaEmbeddings

1 item • Updated 10 days ago • 1

GigaChat3

5 items • Updated 9 days ago • 15

upvoted a paper 11 days ago

Souper-Model: How Simple Arithmetic Unlocks State-of-the-Art LLM Performance

Paper • 2511.13254 • Published 12 days ago • 131

upvoted a collection 12 days ago

MXFP4 Hybrid GGUF

MXFP4 hybrid GGUF models getting... well.. Getting some interesting results. • 11 items • Updated 12 days ago • 3

upvoted a paper 2 months ago

RPG: A Repository Planning Graph for Unified and Scalable Codebase Generation

Paper • 2509.16198 • Published Sep 19 • 127

upvoted a collection 2 months ago

Sep 16 Releases

39 items • Updated Sep 22 • 13

upvoted a paper 2 months ago

DiffusionNFT: Online Diffusion Reinforcement with Forward Process

Paper • 2509.16117 • Published Sep 19 • 21

upvoted an article 3 months ago

Article

Welcome EmbeddingGemma, Google's new efficient embedding model

Sep 4

•

261

upvoted a collection 3 months ago

FastVLM

Efficient Vision Encoding for Vision Language Models • 9 items • Updated Sep 2 • 104

upvoted a paper 3 months ago

Hermes 4 Technical Report

Paper • 2508.18255 • Published Aug 25 • 40

upvoted a collection 3 months ago

Hermes 4 Collection

11 items • Updated about 14 hours ago • 71

upvoted 3 collections 4 months ago

Jamba 1.7

The AI21 Jamba family of models are hybrid SSM-Transformer foundation models, blending speed, efficient long context processing, and accuracy. • 4 items • Updated Jul 2 • 12

GLM 4.5

2 items • Updated Jul 28 • 7

Gemma 3 Release

28 items • Updated Aug 11 • 540

upvoted an article 4 months ago

Article

Fast LoRA inference for Flux with Diffusers and PEFT

Jul 23

•

52