Yujia Qin
@TsingYoga
Followers
5K
Following
249
Media
73
Statuses
334
ByteDancer, Agent, THU (16-20 BS in EE, 20-24 PhD in CS)
Beijing
Joined February 2019
Introducing UI-TARS-1.5, a vision-language model that beats OpenAI Operator and Claude 3.7 on GUI Agent and Game Agent tasks. We've open-sourced a small-size version model for research purposes, more details can be found in our blog. TARS learns solely from a screen, but
18
48
206
RT @deedydas: China's Gaokao is the biggest exam in the world: 13M test takers and 9hrs. ~0.02% make it to the top uni, Tsinghua. As of thi….
0
204
0
Meet Agent TARS Beta, based on Seed1.5-VL.
Since we have released a brand new new Agent TARS CLI based on Seed1.5-VL, see , we have to say goodbye to the old Agent TARS Desktop
2
2
22
RT @_ulivz: Introducing Agent TARS Beta — a brand new and more powerful Agent TARS!. - Agent TARS CLI.- Browser Agent driven by Seed-1.5-VL….
0
16
0
RT @_zhaoheh_: 🚀 UI-TARS Desktop v0.2.1 is now live!.Free Remote Computer & Browser Operator are ready to roll—no setup, just click and go🎁….
0
5
0
RT @_jasonwei: One way of thinking about what AI will automate first is via the “description-execution gap”: how much harder is it to descr….
0
45
0
RT @sainingxie: Had a great time at this CVPR community-building workshop---lots of fun discussions and some really important insights for….
0
61
0
RT @ysu_nlp: @DimitrisPapail I’d argue that computer use, in principle, is much harder than math/coding for current AI. the digital world e….
0
3
0
Guess it's the first open-source multi-turn e2e RL for GUI Agents from academia, and it's based on UI-TARS-1.5-7B. If you want to study multimodal Agent RL, it is a good startpoint~.
9
67
437
The cua community is really amazing!!.
We now have local computer-use! M3 Pro 18GB running both UI-TARS-1.5-7B-6bit and a macOS sequoia VM entirely locally using MLX and c/ua at ~30second/action. Do it yourself here :
0
1
6
Interesting to know the previous operator is based on 4o, not even o1. OpenAI is shifting from reasoning models (o3) to agent models (operator, codex, and deepresearch), with gradual integration of agent data streams from multiple teams—evident in GAIA’s jump from 12.3 to 62.2.
Operator 🤝 OpenAI o3. Operator in ChatGPT has been updated with our latest reasoning model.
3
3
61
RT @giffmana: OK, ByteDance Seed is now firmly a top tier lab in my mind. Congrats on many solid works recently, continuously publishing an….
0
39
0
Interesting findings on the behavior emerging from unified multimodal scaling.
ByteDance just dropped BAGEL on Hugging Face. The Open-Source Unified Multimodal Model
0
3
27