apolinario ๐ŸŒ Profile
apolinario ๐ŸŒ

@multimodalart

Followers
15K
Following
6K
Media
993
Statuses
4K

ML for Art and Creativity, working @HuggingFace ([email protected])

Joined July 2021
Don't wanna be here? Send us removal request.
@Kimi_Moonshot
Kimi.ai
3 hours
๐Ÿš€ Hello, Kimi K2 Thinking! The Open-Source Thinking Agent Model is here. ๐Ÿ”น SOTA on HLE (44.9%) and BrowseComp (60.2%) ๐Ÿ”น Executes up to 200 โ€“ 300 sequential tool calls without human interference ๐Ÿ”น Excels in reasoning, agentic search, and coding ๐Ÿ”น 256K context window Built
214
408
3K
@multimodalart
apolinario ๐ŸŒ
21 hours
Qwen Image Multiple Angles LoRA is an exquisitely trained LoRA! ๐Ÿ“หšโ‚Šโ€ง๊’ฐแƒ Keep character and scenes consistent, and flies the camera around! Open source got there! One of the best LoRAs I've come across lately ๐Ÿ™Œ
49
203
2K
@talkspace
talkspace
20 days
Talkspace online therapy is provided by deeply experienced human therapists using evidence-based methods proven to achieve positive results.
2
11
78
@AnityWolf
Anity Wolf
7 hours
@multimodalart My first try. This totally earned the WOW factor!
0
1
4
@multimodalart
apolinario ๐ŸŒ
21 hours
Qwen Image Multiple Angles LoRA is an exquisitely trained LoRA! ๐Ÿ“หšโ‚Šโ€ง๊’ฐแƒ Keep character and scenes consistent, and flies the camera around! Open source got there! One of the best LoRAs I've come across lately ๐Ÿ™Œ
49
203
2K
@CelsiusOfficial
CELSIUS Energy Drink
2 months
This is more than just four quarters. Itโ€™s every tailgate, every chant, every moment. Itโ€™s fuel that goes beyond the field. This is CELSIUS! LIVE. FIT. GO.
253
422
8K
@linoy_tsaban
Linoy Tsaban๐ŸŽ—๏ธ
1 day
adding camera control to the list of things Qwen Image Edit is great at + with a specialized multi-angle LoRA it's even betterโœจ > rotate the camera > tilt between birdโ€™s-eye and wormโ€™s-eye views > adjust lens (wide, close-up) of course we built a demo for it ๐Ÿค๐Ÿ“น
14
86
662
@wildmindai
Wildminder
4 days
Qwen-Edit 2509 Image Fusion LoRA. Product image composition; blends objects into new backgrounds; automatically attempts to correct perspective and harmonize lighting for seamless integration. https://t.co/5GvrPquRKK
8
61
778
@multimodalart
apolinario ๐ŸŒ
8 days
this is so good! mid-frames are here, multi-frame to video is an easy to use workflow! kudos to @morphic for open sourcing it
@morphic
Morphic
8 days
Morphic's frames-to-video, with up to 5 frames and time control, is now open-source. GitHub: https://t.co/O15v89IaSr Hugging Face: https://t.co/yRDpUni69j More details in the thread:
3
42
314
@multimodalart
apolinario ๐ŸŒ
8 days
Weights are out! ๐Ÿ”ฅ FIBO is a new open weights high quality 8B image model by @bria_ai_ trained on json prompts! It can generate and modify images with a precisely crafted json prompt, allowing for every detail of the image to be decided Try it here: https://t.co/16xNHUGQ8B
3
13
92
@HooverInst
Hoover Institution
15 days
Insights from Condoleezza Rice, Victor Davis Hanson & more. Hoover's top minds tackle policy, security & the economy.
0
49
274
@mariannearr
Marianne Arriola
9 days
๐ŸšจIn our NeurIPS paper, we bring encoder-decoders back.. for diffusion language models! โšก๏ธEncoder-decoders make diffusion sampling fast: a small (fast) decoder denoises tokens progressively and a large (slower) encoder represents clean context.
8
36
240
@multimodalart
apolinario ๐ŸŒ
8 days
this is so good! mid-frames are here, multi-frame to video is an easy to use workflow! kudos to @morphic for open sourcing it
@morphic
Morphic
8 days
Morphic's frames-to-video, with up to 5 frames and time control, is now open-source. GitHub: https://t.co/O15v89IaSr Hugging Face: https://t.co/yRDpUni69j More details in the thread:
3
42
314
@morphic
Morphic
8 days
Morphic's frames-to-video, with up to 5 frames and time control, is now open-source. GitHub: https://t.co/O15v89IaSr Hugging Face: https://t.co/yRDpUni69j More details in the thread:
6
45
332
@multimodalart
apolinario ๐ŸŒ
8 days
@bria_ai_ And the model weights:
Tweet card summary image
huggingface.co
0
0
3
@multimodalart
apolinario ๐ŸŒ
8 days
many folks have been trying json prompting with image and video models, bria trained the model to take that in natively! super cool model ๐Ÿ
@multimodalart
apolinario ๐ŸŒ
8 days
Weights are out! ๐Ÿ”ฅ FIBO is a new open weights high quality 8B image model by @bria_ai_ trained on json prompts! It can generate and modify images with a precisely crafted json prompt, allowing for every detail of the image to be decided Try it here: https://t.co/16xNHUGQ8B
0
2
23
@AustinScholar
Austin Scholar
2 years
Itโ€™s basically a rule of life: teenagers get mad at their parents. But that doesnโ€™t mean you canโ€™t have a relationship with them. If you want to know how to build a connection with your kid (even if theyโ€™re mad at you), read this:
323
480
5K
@multimodalart
apolinario ๐ŸŒ
8 days
Weights are out! ๐Ÿ”ฅ FIBO is a new open weights high quality 8B image model by @bria_ai_ trained on json prompts! It can generate and modify images with a precisely crafted json prompt, allowing for every detail of the image to be decided Try it here: https://t.co/16xNHUGQ8B
3
13
92
@bria_ai_
BRIA AI
8 days
A next-gen visual model trained on structured JSON for precise, controllable generation. ๐Ÿ’ช FIBO is a text-to-image model that transforms prompts into JSON schemas, enabling predictable visuals at scale. Trained on extended, structured captionsโ€”often 1,000+ wordsโ€”FIBO
2
9
52
@Xianbao_QIAN
Tiezhen WANG
8 days
China's open source is just on fire. Soul, China's Tinder (?), has just open sourced their podcast model on @huggingface https://t.co/laie6kUMqd
Tweet card summary image
huggingface.co
16
49
511
@TencentHunyuan
Hunyuan
15 days
Today, we are open-sourcing Hunyuan World 1.1 (WorldMirror), a universal feed-forward 3D reconstruction model. ๐Ÿš€๐Ÿš€๐Ÿš€ ย  While our previously released Hunyuan World 1.0 (open-sourced, lite version deployable on consumer GPUs) focused on generating 3D worlds from text or
45
265
2K
@LivingLibations
Living Libations
6 days
Makeup that works as skincare. Use cosmetic creations crafted with pure plant pigments, essential minerals, and skin-loving lipids.
0
1
1
@xxunhuang
Xun Huang
17 days
Great work scaling Self-Forcing up to 14B models, improving extrapolation, while still keeping it running in real time.
@krea_ai
KREA AI
17 days
Krea Realtime is distilled from the Wan 2.1 14B text-to-video model using Self-Forcing. It achieves a text-to-video inference speed of 11fps using 4 inference steps on a single NVIDIA B200 GPU check all our training methodology and sampling innovations in our technical report!
1
5
112
@krea_ai
KREA AI
17 days
today we're open-sourcing Krea Realtime. this 14B autoregressive model is 10x larger than any open-source equivalent, and it can generate long-form videos at 11 fps on a single B200. weights and technical report below ๐Ÿ‘‡
58
204
1K