ByteDance/Sa2VA-4B

Image-Text-to-Text • 4B • Updated Sep 8 • 170k • • 84

Note A unified model for dense grounded understanding of images & videos.

deepseek-ai/DeepSeek-R1

Text Generation • 685B • Updated Mar 27 • 480k • • 12.8k

Note 660B reasoning models with MIT license

MiniMaxAI/MiniMax-VL-01

Image-Text-to-Text • 456B • Updated Jul 3 • 77.3k • 278

Note A non transformer based ( ViT-MLP-LLM framework) VLM

MiniMaxAI/MiniMax-Text-01

Text Generation • 456B • Updated Jul 3 • 1.33k • 645

Note 456B LLM with 1M tokens training context

Qwen/Qwen2.5-Math-PRM-7B

Text Classification • 8B • Updated Jan 17 • 11.6k • 76

Note Math model

openbmb/MiniCPM-o-2_6

Any-to-Any • 9B • Updated 22 days ago • 98.4k • 1.26k

Note End-side multimodal LLM that supports real time conversation and video understanding.

BlinkDL/rwkv-7-world

Text Generation • Updated May 31 • 104

Note RNN+Transfomers

HKUSTAudio/Llasa-3B

Text-to-Speech • 4B • Updated May 10 • 565 • 523

Note TTS

baichuan-inc/Baichuan-M1-14B-Base

14B • Updated Feb 20 • 55 • 29

Note Medical LLM

Note Dataset designed specifically for natural language processing (NLP) tasks in the education sector.

Note A multimodel dataset for vision language pretraining , includes 6.5M images + 0.8B text from 22k hours of instructional videos