Tag
Qwen
Qwen is Alibaba’s family of open and commercial models spanning text, vision, multimodal, and reasoning variants. It shows up in fine-tuning workflows, math benchmarks, and coding agents, making it a useful reference for performance, cost, and deployment trade-offs.
4 articles
ART fine-tunes multimodal LLMs via pixels
ART fine-tunes frozen multimodal LLMs by optimizing a single input image instead of model weights.
Why Claude Opus 4.8 Is Not the Big Story
Claude Opus 4.8 is less important as a benchmark event than as a sign that model releases are now product updates, not breakthroughs.
QwenPaw Rebrands as a Personal AI Assistant
QwenPaw, formerly CoPaw, rebrands with tighter Qwen ties, local deployment, multi-channel chat, and new v1.1.8 features.
Cursor, Kimi, and the Open Source Bet
Cursor’s new coding model hit near-state-of-the-art performance at one-eighth the price, and its stack points straight to open source.