My goal with the lists are completeness, which is an impossible task. However, the more complete they are—and they are the most complete of anybody here on X—the better Grok will be able to fill in the gaps next year.
LLMS
-
MiniMax-M2.1 Excels on SWE-bench and Long-horizon Tasks
By
–
Excels on SWE-bench (incl. multilingual), Terminal-bench, and long-horizon tasks. Try it on Poe across web, desktop, mobile — and via the Poe API at https://
poe.com/MiniMax-M2.1 (2/2) -
MiniMax M2.1: Agent-Optimized AI Model for Coding and Planning
By
–
Now on Poe: MiniMax M2.1 An agent-optimized model built for coding, tool use, and long-horizon planning served via @novita_labs (1/2)
-
Using Gemini Deep Research for Automated Customer Support Analysis
By
–
Gemini Pro tip: you can use deep research feature to analyze all your emails in Gmail I just used it to analyze all of my customer support interactions from thousands of emails and give me a detailed report of pain points, feature requests, and why people churn
-
User experience comparison between Claude Code and Claude 3.5 Opus
By
–
Arf purée depuis ce tweet je me rend compte que claude code me manque…. . 5.2 c'est vraiment l'IA qui fait un tunnel etc… Sur des projets bien posé il est incroyable mais pour lancer des projets et tester des choses Opus 4.5 est tellement plus agréable. JE SAIS PAS
-
Chinese AI Models Predictions 2026: Open Source Competition
By
–
Here are my 26 predictions for 2026. I tried hard to come up with more spicy predictions that are still plausible – so they sit somewhere in the 5-60% range for me. China 1. Chinese open model tops the Web Dev Arena for 1+ months 2. Chinese labs will open source less than 50%
-
New AI Interaction: Talking to Cognitive Architecture Beyond Prompting
By
–
His cognitive architecture works like the human mind does. And he builds it by talking to it. I watched him work, and he isn't alone, and he showed me the new way of working is not prompting, but talking. And now I agree. It took me a while to "flip the bit" in my head to get
-
Critique of agentic abstraction layers in AI application development
By
–
Non jamais. Je dev tout. Les couche d’abstraction agentique c’est un enfer. J’ai vécu la création et la monté de la hype de langchain et dès le début j’ai voulu en être loin, parce que j’ai connu jusqu’où la pénétration d’une lib à la mode dans une stack peut te flinguer ta dette