SOSOHAJALAB Profile Banner
SUN YOUNG HWANG α―… πŸ‡°πŸ‡· Profile
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·

@SOSOHAJALAB

Followers
727
Following
23K
Media
368
Statuses
7K

CEO, AI (LLM/LMM/MLX) Engineer and enthusiast with non-tech and construction field based. Two kidsπŸ§”β€β™‚οΈ, Fencer🀺, F1 lover🏁 but love tech more πŸ’»β€οΈ

Seoul, Republic of Korea
Joined May 2023
Don't wanna be here? Send us removal request.
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
1 year
Our domain with AI solutions!. - Vision solutions for defective of finished materials object detection with gpt-4-vision and object detection tasks. - Answer the questions for defective of finished materials from GPT, gemini and our own model locally. - OpenAI GPTs for our own
3
0
13
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
1 day
Let me try this!! πŸ‘.
@MistralAI
Mistral AI
2 days
Introducing Devstral Small and Medium 2507! This latest update offers improved performance and cost efficiency, perfectly suited for coding agents and software engineering tasks.
Tweet media one
0
0
0
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
1 day
Another weapon on this?.I might cancel others now πŸ‘
Tweet media one
@alexalbert__
Alex Albert
1 day
We’re rolling out more updates to Claude Code. First, the GitHub Actions integration is now available to Pro and Max plan users (in addition to pay-as-you-go API users).
0
0
0
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
2 days
Another super paper for memory LLM πŸ‘.
@rohanpaul_ai
Rohan Paul
2 days
Brilliant Memory framework proposed in this paper. MemOS makes remembering a first‑class system call. LLMs forget stuff fast and retraining them costs a fortune. MemOS treats memories like files in an operating system, letting a model write, move, and retire knowledge on the
Tweet media one
0
2
9
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
2 days
I might subscribe the grok 4 heavy!. This must be better than claude πŸ‘
Tweet media one
0
0
0
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
2 days
Possible to ship πŸ‡°πŸ‡· πŸ€—.
@ClementDelangue
clem πŸ€—
3 days
Opening orders for Reachy Mini today, our open-source desktop robot for AI builders, starting at $299! Fully integrated with @LeRobotHF & @huggingface for the whole community to build AI apps for it (like this dancing one). We'll probably ship a first batch of a hundred this
0
0
1
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
2 days
RT @AndrewYNg: New Course: Post-training of LLMs. Learn to post-train and customize an LLM in this short course, taught by @BanghuaZ, Assis….
0
330
0
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
2 days
Finally!! Grok 4 is coming real soon πŸ‘.
@elonmusk
Elon Musk
4 days
Grok 4 release livestream on Wednesday at 8pm PT @xAI.
0
0
1
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
2 days
The Godsloth @UnslothAI.
@UnslothAI
Unsloth AI
4 days
We made step-by-step guides to Fine-tune & Run every single LLM! πŸ¦₯. What you'll learn:.β€’ Technical analysis + Bug fixes explained for each model.β€’ Best practices & optimal settings.β€’ How to fine-tune with our notebooks.β€’ Directory of model variants. πŸ”—
Tweet media one
0
0
4
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
3 days
I need this πŸ€—.
@_akhaliq
AK
3 days
Hugging Face just dropped Reachy Mini. an expressive, open-source robot designed for human-robot interaction, creative coding, and AI experimentation. Fully programmable in Python (and soon JavaScript, Scratch) and priced from $299. Reachy Mini measures 11β€œ/28cm in height and
0
0
0
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
3 days
Huggingface has done it always πŸ€—πŸ€—.
@ClementDelangue
clem πŸ€—
3 days
We just released the best 3B model, 100% open-source, open dataset, architecture details, exact data mixtures and full training recipe including pre-training, mid-training, post-training, and synthetic data generation for everyone to train their own. Let's go open-source AI!.
0
0
1
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
3 days
Interesting paper for context consistency πŸ‘.
@omarsar0
elvis
3 days
MemAgent. MemAgent-14B is trained on 32K-length documents with an 8K context window. Achieves >76% accuracy even at 3.5M tokens!. That consistency is crazy!. Here are my notes:
Tweet media one
0
0
1
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
3 days
Another solid method for LLM memory πŸ‘.
@_akhaliq
AK
3 days
MemOS. A Memory OS for AI System
Tweet media one
0
0
0
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
4 days
Vibe coding is still going up πŸ‘.
@minchoi
Min Choi
4 days
Wild. Emergent 2.0 ships full stack apps from a prompt. Now with Google Auth, LLM integration, auto security & scalability reviews. Vibe coding will never be the same. 10 wild examples + how to try:. 1. Veo 3 Ad Wizard
0
0
0
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
5 days
Midjourney is cool as veo3 πŸ€—.
@Ror_Fly
Rory Flynn
5 days
My mom sent me a 30-year-old drawing. That I made for her when I was a kid. Naturally, I animated with Midjourney. It's not perfect. But it's awesome. And it captures the innocent chaos . Of my childhood imagination. Yes, my mother is fighting a dragon. With a
0
0
1
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
7 days
Unstoppable AI models on lmarena πŸ‘.
@lmarena_ai
lmarena.ai
7 days
3 new models live in the Arena today πŸŽ‡ . 🧠 Mistral Small 2506: latest 24B open model (Apache-2.0), tuned for efficiency by @MistralAI.🎨 Imagen 4 Ultra: latest text-to-image from @GoogleDeepMind.πŸ–ŒοΈ Ideogram v3 Quality: latest text-to-image model from @Ideogram_AI. Your votes
Tweet media one
0
0
1
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
7 days
Fantastic job from the Apple πŸ‘.
@ivanfioravanti
Ivan Fioravanti α―…
7 days
Apple released DiffuCoder-7B a Masked Diffusion Model for Code Generation. MLX support is a work in progress right now.
Tweet media one
0
0
1
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
7 days
RT @_akhaliq: Skywork-Reward-V2. Scaling Preference Data Curation via Human-AI Synergy
Tweet media one
0
26
0
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
9 days
Why don't you guys finetune your own datas with good model with THE BEST METHOD @unsloth?! πŸ€—. I've made more than 50models with @unsloth and this is one of our best domain models ever πŸ‘.
@danielhanchen
Daniel Han
9 days
You can utilize our Gemma 3n multimodal and fine-tuning Kaggle notebook for any submission to the $150,000 challenge!. The $10,000 is specifically for the Unsloth track - but you can submit it for the main track as well!. Kaggle notebook:
1
0
2
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
9 days
0
0
0
@SOSOHAJALAB
SUN YOUNG HWANG α―… πŸ‡°πŸ‡·
9 days
Love this mem0 so much!. But, memvid(repoπŸ‘‡(i'm not a contributor πŸ₯²) is really getting better to memory!. This is still improving!. @Gradio can do many things btw πŸ€—
@mem0ai
mem0
9 days
πŸ”§ Easily use Mem0 add and search tools with @OpenAI Agents SDK. - add_memory: saves information to memory.- get_memory: retrieves relevant past information. Persistent memory in your agent workflow in just few lines of code. docs below πŸ‘‡
Tweet media one
1
2
9