@apolinariosteps

(my profile in english is @multimodalart) open source machine learning, IA e criatividade na @huggingface.

 Paris
apolinar.io
Joined May 2009
  • Tweets25.6K
  • Following610
  • Followers2.6K
1.6KMedia
analisei os dados oficiais das urnas do TSE e o voto tardio foi completamente irrisório no resultado eleitoral abri o código pra qualquer um que queira reproduzir a análise github.com/apolinario/urnas-…
2
20
2
50
1,697
é super simples de reproduzir
6
152
just updated the Decision Index to 0.4 🧭 what's new? 🛡️ curbing benchmaxxing: score now exclusively made from private & in-house benchmarks 🔒 stronger private benchmarks on new domains 🏆 224 models evaluated, new top models in all sizes come play 👉huggingface.co/spaces/multim…
16
9
3
55
12,896
the jev Decision Index v0.3 is here 🎯 featuring a way stronger set of benchmarks, now with a strong private half 🔒 so the ranking tracks skills beyond benchmarks and, as promised, vision 👁️ huggingface.co/spaces/multim…
21
26
1
189
18,063
updated Decision Index 0.1 → 0.2 🎯 better formula, +29 jev-like models, +21 benchmarks AutoJev-27B by @perplexity_ai CTO @denisyarats took the open lead, trailing jev by 0.8 points 🏆 come find the best model at every size, speed and use-case huggingface.co/spaces/multim…
22
20
7
175
24,298
apolinário retweeted
Hugging Apps
@HuggingAppsSep 24
Apple just dropped LensVLM 9B on Hugging Face 🍎 a novel kind of VLM that reads long documents the way you skim a book: it glances at every page as a tiny image, then opens only the pages that hold the answer 🔍 ▶️ on Spaces hf.co/spaces/hugging-apps/le…
1
6
1
18
5,645
Introducing the Decision Index 0.1 ⚖️ a rigorous leaderboard comparing jev with 30+ open weights decision models 35+ benchmarks. asking 130K questions to each model testing knowledge 🧠, automation ⚙️, understanding 🤔and even creativity 🎨 huggingface.co/spaces/multim…
28
52
6
311
44,313
apolinário retweeted
Anna Mikeda
@annamikedaSep 18
Round 1 — Send me the 5 bullet points. My first shortcast. Walked SF with @multimodalart, after an hour finally found a reason to fight: is it OK to let AI write your posts? Poli gives a rule for the slop era: if you fed your AI 5 bullet points, just send me those 5 bullet points 😁 Round 2: hyperstition coming soon. 🥊
3
9
507
RT @multimodalart: too many open jev claims and reproductions. which one works? i created this tracker that categorizes 38 artefacts (gith…

Jev Decision Index - a Hugging Face Space by multimodalart

Benchmarks and news on various repros of TypeSafe's Jev

huggingface.co
3
3
apolinário retweeted
Hugging Apps
@HuggingAppsSep 4
StreamTalk is here: audio in, body gesture out, and you get the pose skeleton (SMPL-X), so the 3D rig is up to you 🗣 ▶️ on Spaces hf.co/spaces/hugging-apps/st…
2
8
40
5,091
o verdadeiro uso de IA nas eleições está sendo na geração de jingles e na edição dos vídeos nos programas eleitorais
4
151
Replying to @Merzmensch
That's the spirit of the @MediaSynthesisM We have Big Sleep, VQGAN+CLIP, CLIP Guided Diffusion and some other models all preserved and restored to run on modern hardware huggingface.co/mediasynthesi…

mediasynthesismuseum (Media Synthesis Museum)

Org profile for Media Synthesis Museum on Hugging Face, the AI community building the future.

huggingface.co
5
2
14
434
chuvas malucas em brasília trazendo de volta o apolinar.io/jachoveu/
yaya é PSOL e PSOL é 1️⃣3️⃣ 🇵🇸@yayaloureiroAug 25
Tem um site que monitora em quais anos choveu em cada dia do ano desde a fundação de Brasília. A chuva de hoje apesar de incomum já havia aparecido em alguns anos mais recentes. Mas a de sábado passado foi mais surpreendente pois nessa data só havia 40 anos que não chovia
2
2
44
1,952
apolinário retweeted
Hugging Apps
@HuggingAppsAug 21
SenseNova-U1.5-8B-MoT is HERE, it's a big deal! 🔥 open weights matching Nano Banana 2 on benchs 🍌 no VAE, no text encoder, no DiT. a mixture of transformers: text and image tokens use different weights, attend to each other, denoise in pixel space ▶️ huggingface.co/spaces/huggin…
4
55
2
542
42,957
apolinário retweeted
Hugging Apps
@HuggingAppsAug 19
unmute any clip! Xiaomi just dropped ControlFoley on Hugging Face 🎬🔇→ 🔊 a video-to-audio model that generates audio based on what's happening on the video. prompt optional. audio reference guidance supported ▶️ on Spaces huggingface.co/spaces/huggin…
13
10
1
111
6,091
apolinário retweeted
Hugging Apps
@HuggingAppsAug 14
find & replace, but for a voice recording 🎙️ FireRedTTS3 is out on @huggingface, an omni-TTS model that can do voice design, voice cloning and a cool new feature: speech in-painting - or voice editing! ▶️ on Spaces huggingface.co/spaces/huggin…
8
38
3
290
15,283
MiniMax Music 3 is just out on @huggingface 🎶🎵 SoTA song generation, now open weights 🔊, a 8B LLM + 2.7B DiT that turn prompt + lyrics into full songs fits smol consumer GPUs with 🧨 diffusers or comfy model weights + app on Spaces, come play ▶️ huggingface.co/spaces/MiniMa…
16
44
14
420
64,315
apolinário retweeted
Hugging Apps
@HuggingAppsAug 9
extremely precise control just landed to Krea-2 as a LoRA! 🎨 krea2-turbo-bbox adds precise layout-controlled generation to Krea 2: drag the boxes, regenerate, the composition obeys, text boxes included cooked by jimmycarter 🧑‍🍳 ▶️ on Spaces huggingface.co/spaces/huggin…
10
50
6
475
28,371
apolinário retweeted
Hugging Apps
@HuggingAppsAug 7
ScenA is a new omni-TTS model out on Hugging Face, it takes a text description + two voice references and produces a full mixed scene with sound effects 🎧 does pineapple belong on pizza? 🍕 🍍 ▶️ on Spaces huggingface.co/spaces/huggin…
3
26
2
237
22,980
apolinário retweeted
Hugging Apps
@HuggingAppsAug 1
CharacterSheet LoRA by @Alissonerdx just dropped on HF and is so good 🎨 feed it one image of a character, photo or illustration, and this FLUX.2-klein LoRA generates a full model sheet: portrait, front, side and back, identity intact ▶️ on Spaces huggingface.co/spaces/huggin…
9
34
299
16,604
Load more