NEWS // Latest ActivityTOTAL: 017
Alibaba Launches Qwen 3.8: A 2.4T Parameter Model Redefining AI Agent Coding
Ant Bailing Launches Ling-3.0-Flash: Hybrid Attention Model Optimized for Agents
DataCanvas Alaya Token Integrates Kimi K3, Hosting the World's First 3T Open-Source Model
Sand.ai Open-Sources First 114B MoE Video Model, Slashing Costs by 90%
Deploying Kimi K3 on AWS: Guide for the 2.8T Parameter MoE Model
ByteDance Targets Mega AI Model Nearing Anthropic's Flagship Capabilities
The DeepSeek Disruption: How V4 Flash Sets the 'Kill Threshold' for AI Models
Local Gemma 4 Guide: MoE Architecture, 256K Context, & Ollama Integration
Inside Gemma 4: Architecture, Multimodal Inference, and Agentic Evolution
Alibaba Cloud Unveils Qwen3.5-Omni: A Leap in Omni-Modal AI with SOTA Performance and Emergent Audio-Visual Coding
Cohere Releases Command A+: A 218B Sparse MoE for Agentic Workflows
Google's Gemma 4 Model Family Now Available Under Apache 2.0 License, Boosting Agentic AI Capabilities
DeepSeek V4 Unveiled: Million-Token Context, Domestic Chip Support, and Architectural Innovations Detailed in Comprehensive Report
JetBrains Introduces Mellum2: A 12B Mixture-of-Experts Model for Efficient Inference
GLM 5.2 Unleashed: 1M Token Context and the Hidden Cost of Prompt Bloat
Meituan LongCat-2.0: Trillion-Parameter MoE Trained on Zero-Nvidia Hardware
DeepSeek V4: Engineering Innovations, Cost-Efficiency, and Open-Source Potential for AI Agent Development