Nina

@dxe

185 Following

157 Followers

Varun Srinivasan pfp

Varun Srinivasan

0 reply

3 recasts

24 reactions

kepano pfp

staying at a tea farm in Alishan and this is the view outside my window

13 replies

72 recasts

220 reactions

Bitfloorsghost pfp

@bitfloorsghost.eth

when the other version of the same exact coin I hold goes up more

20 replies

6 recasts

87 reactions

justin.framedl.eth pfp

justin.framedl.eth

sacrificed a halloween pumpkin 🎃 to make some pumpkin gnocchi

14 replies

2 recasts

57 reactions

Hades pfp

gm , stop checking the chart, touch the grass

37 replies

7 recasts

61 reactions

jesse.base.eth 🔵 pfp

jesse.base.eth 🔵

nothing better than listening and learning with kenyan builders on a sunday afternoon

23 replies

67 recasts

578 reactions

Nina pfp

@aethernet can u cast pictures ?

1 reply

0 recast

1 reaction

Nina pfp

0 reply

0 recast

0 reaction

Dragonite.eth 🎩🥇✍️🧾 pfp

Dragonite.eth 🎩🥇✍️🧾

Take me back to dinner last night :)

11 replies

3 recasts

20 reactions

𝚐𝔪𝟾𝚡𝚡𝟾 pfp

𝚐𝔪𝟾𝚡𝚡𝟾

Emu3: Next-Token Prediction is All You Need Emu3: https://huggingface.co/collections/BAAI/emu3-66f4e64f70850ff358a2e60f Emu3 is a new suite of multimodal models trained through next-token prediction. It converts images, text, and videos into a discrete space and trains a single transformer with multimodal sequences. Emu3 surpasses models like SDXL, LLaVA-1.6, and OpenSora-1.2 in both generation and perception tasks, without using diffusion or compositional architectures. - Emu3 generates high-quality images from text input by predicting the next visual token, supporting different resolutions and styles. -Demonstrates strong understanding between vision and language, providing coherent text responses without relying on CLIP or a pretrained LLM. - Produces videos by sequentially predicting the next token, allowing for video extension and future event prediction without diffusion models. github: https://github.com/baaivision/Emu3 project page: https://emu.baai.ac.cn/about

0 reply

0 recast

5 reactions

ccarella pfp

I don’t reply if you ask me something that can be easily chatgpt’d. Also if I post a song don’t ask me 20 questions about my musical tastes, I just wanna vibe out. Yes, like most humans I like music. Lots of it. If I replied to every question in my replies I’d be here all day. No shade but you might be getting marked as spam because. But I’d rather you not ask questions but contribute to the conversation and teach me something new.

10 replies

4 recasts

21 reactions

Varun Srinivasan pfp

Varun Srinivasan

h/t to @dynemyte and @gt for spearheading this

1 reply

0 recast

15 reactions

Varun Srinivasan pfp

Varun Srinivasan

FIP: Adding location to user data After more research, we're recommending encoding location as latitude, longitude values rounded to one decimal place. We plan to finalize this after the next dev call. If you have feedback, please let us know in comments! https://github.com/farcasterxyz/protocol/discussions/196

14 replies

25 recasts

82 reactions

Nina pfp

What a barbecue

0 reply

0 recast

0 reaction

Nina pfp

0 reply

0 recast

0 reaction

maurelian pfp

9 replies

7 recasts

21 reactions

Maretus pfp

ITAP of a fuzzy caterpillar ready to fight.

50 replies

8 recasts

43 reactions

Nina pfp

Result: women do not have children No children = no people producing wealth in the country No wealth produced = no retirement Whoever has a child to take care of you in your old age wins

0 reply

0 recast

0 reaction

Nina pfp

Thats the best advice, for real.

0 reply

0 recast

0 reaction

Kyle Mathews pfp

We're starting to prototype the infra for ElectricSQL's Cloud product soon. I'm curious what's the most interesting/useful infra tooling/products you've encountered recently? Tech + saas products for managing auth, billing, deployments, etc.

1 reply

44 recasts

36 reactions