Kai Ruan's picture

Kai Ruan

6cf

·

x66ccff

AI & ML interests

AI for Science

Recent Activity

reacted to merve's post with 🧠 about 6 hours ago

Oof, what a week! 🥵 So many things have happened, let's recap! https://huggingface.co./collections/merve/jan-24-releases-6793d610774073328eac67a9 Multimodal 💬 - We have released SmolVLM -- tiniest VLMs that come in 256M and 500M, with it's retrieval models ColSmol for multimodal RAG 💗 - UI-TARS are new models by ByteDance to unlock agentic GUI control 🤯 in 2B, 7B and 72B - Alibaba DAMO lab released VideoLlama3, new video LMs that come in 2B and 7B - MiniMaxAI released Minimax-VL-01, where decoder is based on MiniMax-Text-01 456B MoE model with long context - Dataset: Yale released a new benchmark called MMVU - Dataset: CAIS released Humanity's Last Exam (HLE) a new challenging MM benchmark LLMs 📖 - DeepSeek-R1 & DeepSeek-R1-Zero: gigantic 660B reasoning models by DeepSeek, and six distilled dense models, on par with o1 with MIT license! 🤯 - Qwen2.5-Math-PRM: new math models by Qwen in 7B and 72B - NVIDIA released AceMath and AceInstruct, new family of models and their datasets (SFT and reward ones too!) Audio 🗣️ - Llasa is a new speech synthesis model based on Llama that comes in 1B,3B, and 8B - TangoFlux is a new audio generation model trained from scratch and aligned with CRPO Image/Video/3D Generation ⏯️ - Flex.1-alpha is a new 8B pre-trained diffusion model by ostris similar to Flux - tencent released Hunyuan3D-2, new 3D asset generation from images

reacted to merve's post with 👍 about 6 hours ago

Oof, what a week! 🥵 So many things have happened, let's recap! https://huggingface.co./collections/merve/jan-24-releases-6793d610774073328eac67a9 Multimodal 💬 - We have released SmolVLM -- tiniest VLMs that come in 256M and 500M, with it's retrieval models ColSmol for multimodal RAG 💗 - UI-TARS are new models by ByteDance to unlock agentic GUI control 🤯 in 2B, 7B and 72B - Alibaba DAMO lab released VideoLlama3, new video LMs that come in 2B and 7B - MiniMaxAI released Minimax-VL-01, where decoder is based on MiniMax-Text-01 456B MoE model with long context - Dataset: Yale released a new benchmark called MMVU - Dataset: CAIS released Humanity's Last Exam (HLE) a new challenging MM benchmark LLMs 📖 - DeepSeek-R1 & DeepSeek-R1-Zero: gigantic 660B reasoning models by DeepSeek, and six distilled dense models, on par with o1 with MIT license! 🤯 - Qwen2.5-Math-PRM: new math models by Qwen in 7B and 72B - NVIDIA released AceMath and AceInstruct, new family of models and their datasets (SFT and reward ones too!) Audio 🗣️ - Llasa is a new speech synthesis model based on Llama that comes in 1B,3B, and 8B - TangoFlux is a new audio generation model trained from scratch and aligned with CRPO Image/Video/3D Generation ⏯️ - Flex.1-alpha is a new 8B pre-trained diffusion model by ostris similar to Flux - tencent released Hunyuan3D-2, new 3D asset generation from images

reacted to merve's post with 👀 about 6 hours ago

Oof, what a week! 🥵 So many things have happened, let's recap! https://huggingface.co./collections/merve/jan-24-releases-6793d610774073328eac67a9 Multimodal 💬 - We have released SmolVLM -- tiniest VLMs that come in 256M and 500M, with it's retrieval models ColSmol for multimodal RAG 💗 - UI-TARS are new models by ByteDance to unlock agentic GUI control 🤯 in 2B, 7B and 72B - Alibaba DAMO lab released VideoLlama3, new video LMs that come in 2B and 7B - MiniMaxAI released Minimax-VL-01, where decoder is based on MiniMax-Text-01 456B MoE model with long context - Dataset: Yale released a new benchmark called MMVU - Dataset: CAIS released Humanity's Last Exam (HLE) a new challenging MM benchmark LLMs 📖 - DeepSeek-R1 & DeepSeek-R1-Zero: gigantic 660B reasoning models by DeepSeek, and six distilled dense models, on par with o1 with MIT license! 🤯 - Qwen2.5-Math-PRM: new math models by Qwen in 7B and 72B - NVIDIA released AceMath and AceInstruct, new family of models and their datasets (SFT and reward ones too!) Audio 🗣️ - Llasa is a new speech synthesis model based on Llama that comes in 1B,3B, and 8B - TangoFlux is a new audio generation model trained from scratch and aligned with CRPO Image/Video/3D Generation ⏯️ - Flex.1-alpha is a new 8B pre-trained diffusion model by ostris similar to Flux - tencent released Hunyuan3D-2, new 3D asset generation from images

View all activity

Organizations

None yet

6cf's activity

upvoted a paper 2 days ago

Discovering symbolic expressions with parallelized tree search

Paper • 2407.04405 • Published Jul 5, 2024 • 1

upvoted a collection 5 days ago

DeepSeek-R1

8 items • Updated 6 days ago • 148

upvoted a collection 19 days ago

IdeaWhiz

3 items • Updated 17 days ago • 3

upvoted a collection 22 days ago

DeepSeek-V3

3 items • Updated 21 days ago • 135

upvoted a collection 25 days ago

Models for Open Hands (Open Devin)

Models for Open Hands(Open Devin) trained on the Devinator Dataset. • 3 items • Updated Oct 25, 2024 • 2

upvoted a collection 28 days ago

LiveIdeaBench

4 items • Updated 18 days ago • 5

upvoted a paper 29 days ago

LiveIdeaBench: Evaluating LLMs' Scientific Creativity and Idea Generation with Minimal Context

Paper • 2412.17596 • Published Dec 23, 2024 • 6

upvoted 2 collections about 1 month ago

YuLan-Mini

A highly capable 2.4B lightweight LLM using only 1T pre-training data with all details. • 5 items • Updated 28 days ago • 11

DeepSeek-VL2

4 items • Updated Dec 18, 2024 • 37

upvoted an article about 2 months ago

Article

🐺🐦‍⬛ LLM Comparison/Test: 25 SOTA LLMs (including QwQ) through 59 MMLU-Pro CS benchmark runs

By

•

Dec 4, 2024

• 76

upvoted a collection about 2 months ago

Recommended large models

This collection contains some of the recent models larger than ~25B parameters that should be high quality and reliable • 15 items • Updated Nov 27, 2024 • 13

upvoted a collection 4 months ago

Qwen2.5-Coder

Code-specific model series based on Qwen2.5 • 40 items • Updated Nov 28, 2024 • 262