Qwen 2.5, every mention
3 scenes · ← back to Qwen 2.5
tap a year for its mentions
every year anyone Pratyush Maini 1Loubna Ben Allal 1
Verbatim, from the transcripts: the passages where Qwen 2.5 comes up
⚡️ Reverse Engineering OpenAI's Training Data — Pratyush Maini, Datology
- ▶ 4:59 Pratyush Maini I don't, we don't see this phenomenon like the earlier versions of like QN 1.5 or even QN two, but start seeing it in QN three.
[State of AI Papers 2025] Fixing Research with Social Signals, OCR & Implementation — Team AlphaXiv
- ▶ 19:22 unnamed speaker And so, you know, when you've applied that framework on top of, you know, a base model, like one, 2.5, you suddenly now have an agent that is very, very good at these sort of, these more complex tasks. 3 times in the scene
Best of 2024: Synthetic Data / Smol Models, Loubna Ben Allal, HuggingFace [LS Live! @ NeurIPS 2024]
- ▶ 17:28 Loubna Ben Allal For example, if you look at this, um, chart from the QN 2.5 blog posts, it shows that today we have some really good models that are only like three billion parameters in four billion, the score really high on MMLU, which is a really…