Radamés Ajna's picture

Radamés Ajna

radames

·

https://observablehq.com/@radames

AI & ML interests

None yet

Recent Activity

reacted to cbensimon's post with 👍 2 days ago

🚀 ZeroGPU `medium` size is now available as a power-user feature Nothing too fancy for now—ZeroGPU Spaces still default to `large` (70GB VRAM)—but this paves the way for: - 💰 size-based quotas / pricing (`medium` will offer significantly more usage than `large`) - 🦣 the upcoming `xlarge` size (141GB VRAM) You can as of now control GPU size via a Space variable. Accepted values: - `auto` (future default) - `medium` - `large` (current default) The auto mode checks total CUDA tensor size during startup: - More than 30GB → `large` - Otherwise → `medium`

reacted to cbensimon's post with 👀 2 days ago

🚀 ZeroGPU `medium` size is now available as a power-user feature Nothing too fancy for now—ZeroGPU Spaces still default to `large` (70GB VRAM)—but this paves the way for: - 💰 size-based quotas / pricing (`medium` will offer significantly more usage than `large`) - 🦣 the upcoming `xlarge` size (141GB VRAM) You can as of now control GPU size via a Space variable. Accepted values: - `auto` (future default) - `medium` - `large` (current default) The auto mode checks total CUDA tensor size during startup: - More than 30GB → `large` - Otherwise → `medium`

reacted to cbensimon's post with 🤗 2 days ago

🚀 ZeroGPU `medium` size is now available as a power-user feature Nothing too fancy for now—ZeroGPU Spaces still default to `large` (70GB VRAM)—but this paves the way for: - 💰 size-based quotas / pricing (`medium` will offer significantly more usage than `large`) - 🦣 the upcoming `xlarge` size (141GB VRAM) You can as of now control GPU size via a Space variable. Accepted values: - `auto` (future default) - `medium` - `large` (current default) The auto mode checks total CUDA tensor size during startup: - More than 30GB → `large` - Otherwise → `medium`

View all activity

Organizations

radames's activity

upvoted an article 3 days ago

Article

Tiny Agents in Python: a MCP-powered agent in ~70 lines of code

By

and 3 others •

7 days ago

• 95

upvoted a collection 14 days ago

VACE

VACE: All-in-One Video Creation and Editing • 7 items • Updated 15 days ago • 24

upvoted a collection 18 days ago

FramePack Video Generation

fast & compact video generation with FramePack - a next-frame prediction neural network structure that generates videos progressively • 9 items • Updated 4 days ago • 5

upvoted 5 collections about 1 month ago

HiDream-I1

A collections of HiDream-I1 models. • 4 items • Updated Apr 8 • 31

HiDream-E1

A collections of HiDream-E1 models. • 2 items • Updated Apr 28 • 3

Qwen3

Qwen's new Qwen3 models. In Unsloth Dynamic 2.0, GGUF, 4-bit and 16-bit Safetensor formats. Includes 128K Context Length variants. • 65 items • Updated about 11 hours ago • 145

Perception LM

7 items • Updated Apr 17 • 52

ReflectionFlow release

https://diffusion-cot.github.io/reflection2perfection/ • 6 items • Updated Apr 23 • 10

upvoted an article about 1 month ago

Article

Tiny Agents: a MCP-powered agent in 50 lines of code

By

•

Apr 25

• 263

upvoted 2 papers 2 months ago

InfiniteYou: Flexible Photo Recrafting While Preserving Your Identity

Paper • 2503.16418 • Published Mar 20 • 35

SANA-Sprint: One-Step Diffusion with Continuous-Time Consistency Distillation

Paper • 2503.09641 • Published Mar 12 • 39

upvoted 2 articles 3 months ago

Article

Remote VAEs for decoding with HF endpoints 🤗

By

and 1 other •

Feb 24

• 39

Article

SigLIP 2: A better multilingual vision language encoder

By

and 2 others •

Feb 21

• 164

upvoted a collection 4 months ago

Reasoning Datasets

Distilled synthetic Reasoning datasets • 7 items • Updated Feb 2 • 61

upvoted an article 4 months ago

Article

Welcome to Inference Providers on the Hub 🔥

By

and 6 others •

Jan 28

• 479

upvoted 2 collections 4 months ago

SAEs

Sparse Autoencoders trained by qresearch • 3 items • Updated Jan 29 • 2

Contra (Bottleneck T5)

Text autoencoders capable of embedding and generating text in a fixed-size latent space, useful for embeddings and latent space text editing. • 4 items • Updated Oct 3, 2023 • 28

upvoted an article 4 months ago

Article

The AI tools for Art Newsletter - Issue 1

By

and 1 other •

Jan 31

• 78

upvoted a paper 4 months ago

DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Paper • 2501.12948 • Published Jan 22 • 392

upvoted a collection 5 months ago

DeepSeek-V3

4 items • Updated Mar 25 • 257