LLMS

→ View original post on X — @linusekenstam

LTX-2: Open-Source Model with 3.3M Downloads in One Month

By

@linusekenstam

–

14 February 2026 20h39

https://t.co/BbJXl2NWd7
— Linus ✦ Ekenstam (@LinusEkenstam) 14 février 2026

+3.3 Million downloads in 1 Month Yet most people are sleeping on LTX-2 plus the fact that its Open-Source
That's right, you can even run it at home, at the cost of electricity. Let's dig into why I think you should spend some time with this model. A thread

→ View original post on X — @dair_ai

Improving Efficiency of Evolutionary AI Agents

By

@dair_ai

–

14 February 2026 15h48

// Improving Efficiency of Evolutionary AI Agents // Evolutionary AI agents are powerful but can be wasteful. Systems, inspired by AlphaEvolve and OpenEvolve, iteratively generate, mutate, and refine candidate solutions using LLMs. However, every refinement step invokes the

Evaluation of Claude Opus as an assistant model

By

–

14 February 2026 12h57

Yeah opus is def the better assistant model

Reflecting on GPT-4o as a personality-driven AI assistant

By

–

14 February 2026 12h56

4o is the personality hire we keep around

User feedback on Claude Opus behavior and bias

By

–

14 February 2026 12h55

Claude opus being over eager and biased towards action is so true too

Practical AI workflow for coding with Opus and Codex

By

–

14 February 2026 12h53

Interestingly I like vibe coding stuff with opus then bringing in codex to refactor and build further – haven’t had issues yet

→ View original post on X — @julien_c

LanceDB and Hugging Face Partner for Vector Storage

By

@julien_c

–

14 February 2026 12h24

in case you missed it @lancedb and HF are partnering up to unlock the next generation of large dataset storage on the Hub And it's fire! – Supports storing embeddings (and their indexes) directly alongside the data
– Vector search / similarity search is built-in
– Large