3d ago · 6 min read · So, instead of the usual models that use all their settings when making predictions, Qwen3.5-122B-A10B has a cool setup called Mixture-of-Experts (MoE). This allows the model to activate only a small
Join discussion
3d ago · 9 min read · Qwen3-Coder-Next is one of the most compelling entries in this new generation of developer-focused models. Developed by Alibaba's Qwen team, it is an open-weight MoE language model designed specifical
Join discussion
3d ago · 12 min read · But if you’re actually building with these models, the real question is much simpler: What happens when you give them the same prompt and ask them to write code? So we decided to test exactly that usi
Join discussion
Sep 29, 2025 · 3 min read · Introduction In 2025, Alibaba Cloud (阿里云) announced and released new Qwen3 (通义千问3) large language models (LLM), which is part of their Qwen family of LLMs. As of 29 September 2025, Alibaba Cloud released following Qwen3 models: Qwen3-235B-A22B A la...
Join discussionAug 31, 2025 · 4 min read · This document outlines the complete, step-by-step process for deploying the Qwen/Qwen3-Coder-30B-A3B-Instruct model on a DigitalOcean H100 droplet. The final setup uses vLLM for high-performance inference, Nginx as a secure reverse proxy for API key ...
Join discussion
Aug 21, 2025 · 4 min read · There’s a quiet shift happening in developers' workspaces. It’s not in the hardware or a new programming language; it’s in the soft glow of the code editor itself. In 2025, AI-powered coding tools will have evolved from neat party tricks into indispe...
Join discussionAug 10, 2025 · 1 min read · TLDR - Rovo dev CLi by Atlassian. Why? Based on Claude Code Uses Claude Sonnet 4. Update: they also just started giving a choice between Sonnet and GPT-5 20 Million tokens per 24 hours in free usage That’s the post. You wanted more? I will give...
Join discussionJul 24, 2025 · 3 min read · Introduction The push for accessible, high-performance code assistants has brought us Qwen3-Coder. This open source model offers an enormous 1M token context window and a unique architecture that positions it as a strong alternative to established pr...
Join discussion
