Tech news & AI articles - LinkWord
Needle: Distilling Gemini's Tool Calling into a 26M Parameter Model
Cactus Compute distills Gemini 3.1's tool calling into Needle, a 26M parameter model that beats models 10x its size on function calling. Open source with 2100+ GitHub stars.
5/18/2026
4 views
I Turned Minecraft Server Operations into a Claude Code Skill - Now Open Source
A Claude Code Skill covering both Java and Bedrock Edition Minecraft servers, handling everything from setup to troubleshooting. Includes an RCON client built with pure Python standard library, requiring zero external dependencies.
5/16/2026
14 views
Kimi K2.6 Officially Released: Major Breakthroughs in Long-Context Coding and Multi-Agent Cluster Capabilities
Moonshot AI officially released and open-sourced its latest flagship model, Kimi K2.6, on April 20. The model achieved significant upgrades in code generation, long-horizon task execution, and multi-agent cluster capabilities, performing on par with or surpassing mainstream closed-source models like GPT-5.4 and Claude Opus 4.6 across multiple benchmarks. This marks a new milestone for domestic large models in complex software engineering and automated task processing.
4/21/2026
13 views
Alibaba Qwen3.6: China's Leading Coding Model Open Sourced, with Comprehensive Agent Capability Upgrades
In April 2026, Alibaba Tongyi Qianwen unveiled the next-generation Qwen3.6 large language model series, delivering significant breakthroughs in coding proficiency and agent performance. Following the launch of the flagship Qwen3.6-Plus on April 2, the team open-sourced the lightweight and efficient Qwen3.6-35B-A3B model on April 16. This milestone signifies that domestic large models have achieved global leadership in the coding domain.
4/17/2026
22 views
MixMax-2.7 Open Source Project In-depth Analysis: A Comprehensive Comparison with GPT-5.4 and GLM-5
In 2026, the open-source wave of large models reached a critical juncture. MiniMax's open-source M2.7 (often mistakenly referred to as MixMax-2.7) redefined the AI development paradigm with its "self-evolving" agents; OpenAI's GPT-5.4 integrated native computer operation capabilities for the first time, pushing AI into professional work scenarios; while Zhipu's GLM-5 topped the open-source charts with its "agent engineering" philosophy. This article provides an in-depth analysis of these three landmark models from multiple dimensions—technical architecture, core capabilities, open-source ecosystem, and application scenarios—offering crucial reference for developers and enterprises in selecting their AI foundation.
4/14/2026
21 views