Dango233
@dango233max
Followers
900
Following
174
Media
29
Statuses
142
Baking open AI system Garnishing open weights https://t.co/Y5yWy3Hn2K
Shenzhen
Joined November 2011
Find research faster with II-Commons Search arXiv + PubMed (web app + Agent demo, API/MCP/A2A) Beta now live for II-Accounts
7
13
55
While building II’s open stack, we put together a Gemini-CLI → MCP+OpenAI a bridge to access the tools + model in our tests. This lets any MCP-savvy agent tap Gemini + its tools via your gemini-cli instance!
6
21
78
II-Medical-8B-1706 is our latest state of the art open medical model 💡 Outperforms the latest @Google MedGemma 27b model with 70% less parameters 🤏 Quantised GGUF weights, works on <8 Gb RAM 🚀 One more step to the universal health knowledge access that everyone deserves ⚕️
21
119
587
II-Commons Infrastructure for shared knowledge. Transparent. Distributed. Open source. The foundation for trustworthy AI.
1
21
75
Thank you for choosing SD3.5M as the training base! This is exactly the kind of research we hoped for this Medium (kinda small by today's standard...) sized model!!
Thanks for sharing your work! If your generative model uses flow-matching, feel free to try our Flow-GRPO. Our code is at:
0
1
4
II-Researcher sets a new high on the @GoogleDeepMind FRAMES RAG benchmark with 80% accuracy. Read the in-depth blog post, use the open source code & try it out yourself.
8
26
106
♊ Introducing II-Researcher: Our open-source AI research agent with strong benchmark performance ♊ 🚀 Key highlights: • New open-source framework for complex, multi-step search with reasoning tasks— a powerful, transparent alternative to proprietary systems • Two methods
29
117
644
QwQ-32b Is Indeed The World's Best Open Source Model. We re-ran QwQ with the suggested settings from the Qwen team, and it turns out that it's an AMAZING LLM It's got excellent scores on Livebench AI.
91
154
1K
CoT-Lab中文界面的API提供方切到硅基流动了，丝滑～ 这是一个实验性质的，探索和reasoning模型交互方式的界面，侧重在一轮对话中跟模型共同迭代一个思维链。 - 修改思维过程。 做了一些交互优化，希望大家能从跟模型共创思维的过程中找到一些特别的体验。
4
1
9
我们做了一个很有意思的东西。你可以用它来调整R1输出的速度，从而让自己的思考和r1的<think>同步，可以随时暂停修正你觉得R1思考错误的地方，把一些你的思想"植入"R1的思考过程中，最后得到最好的结果。似乎是R1特殊的训练方式，在其他reasoning模型上都很难做到这个。大家可以来玩玩。
DeepSeek R1 is great. How do humans think with reasoning machines like R1? CoT-Lab represents @ii_posts's latest exploration at UBAI into cognitive partnership — where human intuition and AI reasoning become co-evolving thought partners. We enable collaboration with reasoning
14
86
422
https://t.co/c7mNoZbYa3 HF space could be unstable time-to-time due to unstable API. Clone it and run with your local hosted R1-Distilled, or visit for the dev codebase https://t.co/jgbWoQAjnw
github.com
Sync your thinking with AI reasoning models to achieve deeper cognitive alignment Follow, learn, and iterate the thought within one turn - Intelligent-Internet/CoT-Lab-Demo
0
0
0
When your thought partner shares their mind, do you: - Follow their path? - Learn from their wisdom? - Or... reshape the ideas together? CoT-Lab is an experiment looking into Human-AI cognitive colaboration. Thanks to @deepseek_ai , we have a thought partner to work with!
DeepSeek R1 is great. How do humans think with reasoning machines like R1? CoT-Lab represents @ii_posts's latest exploration at UBAI into cognitive partnership — where human intuition and AI reasoning become co-evolving thought partners. We enable collaboration with reasoning
1
1
3
Anyone who thinks DeepSeek just came out of nowhere should see this graph. For each model on this graph, weights, code, and detailed papers were released. This is a team with a strong track record and has been working hard for a while. They didn't come out of nowhere.
74
650
4K