Hugging Face Releases Carbon-X, a 14B Open DNA Language Model that Brings DNA and Natural Language Together for the First Time

Nanbeige Releases Nanbeige4.3-33B, a 33B Open Agentic Coding Model

Researchers from Stanford and UC Berkeley Introduce TokenGate: Token-Gated Attention Makes Long-Context Prefill Up to 11.3x Faster

Microsoft Open-Sources Agensh: A Multi-Agent Harness for 1,024 Agents

Anthropic Launches OSS Scanner, a Free AI Vulnerability Scanner

How to Run Underdog Saluki 27B: 2-Bit Qwen3.8-27B Tool-Calling Guide

Cognition Open-Sources an Agent Plugin that Helps Secure Systems against Recent Cybersecurity Threats

Meet Anthropic’s sandbox-runtime: OS-Level Sandboxing for AI Agents, No Containers

Stateful Agentic Engineering Copilots with Letta: Core Memory, Shared State, Semantic Retrieval, Custom Tools, Streaming, and Portable Agent Workflows

Moonshot AI Releases Kimi K3.1 Pro: A 2.8T Total/104B Active Params (MoE) Model with 1M Context

Kyutai CADRE: Video Autoencoders That Cut World Model Lag Below 230 ms

NVIDIA Releases Nemotron 3.5 Bolt E4B 4GB: An Open Small Language Model for Local Voice Agents, Game Characters and Robots

Gradium Open-Sources Phonon, a 31M-Parameter On-Device TTS Model Under Apache 2.0

Google Cloud Launches Agent Compute Engine: A MicroVM per AI Agent

TinyFish Breaks Down Agentic Search Architecture for Production AI Agents

Liquid AI Launches RIPL, a Real-Time Inference Engine for Sub-50 ms LFMs

You.com Launches Inference Router: 12 Open-Weight Models With 99.9% Uptime SLA

NVIDIA Releases Nemotron 3.5 Super VL, a 120B Open Multimodal Agent Model

Fish Audio Releases Drama 3, a 35B Open-Weight MoE Text-to-Speech Model

Datalab Releases OmniParseBench: An Open OCR Benchmark With 16,288 Pass/Fail Tests to Audit OCR Parsers

Meet Grok Voice 3.0: SpaceXAI Voice Agent Model Ranks #1 on Artificial Analysis

TypeSafe AI Releases Jev 1.14 With Images, Box Type and 262k Context

openKylin Outlines Its Vision for Reimagining Linux in the AI Era at Open Source Summit Europe 2026

CopilotKit Releases CLiF: A Continual Learning Platform That Turns Agent Traces and User Actions into Skills, Evals, and Fine-Tuning Data

Liquid AI Releases LFM2.5-24B-A2B: A 2B-Active MoE for On-Device Agents

Google Releases Gemini 4 Argon API With 1M Token Output Limit

9 Best AI Gateways in 2026: Bifrost, Portkey, LiteLLM, Kong and 5 More Ranked

Alibaba Qwen Introduces QwenGyre: Elastic Reinforcement Learning RL for XLong-Horizon Agents

What is Decision 3.0? vLLM Semantic Router’s New Open Decision Models

Meet Cloudflare Clef-omni: A 30B-A3B Multimodal Decision Model for Audio, Video, Image and Text

Top 12 AI Live Translation Tools in 2026: Latency, Languages, Pricing

TII Releases Falcon-ASR: 1.6B Arabic Speech Model Hits 20.92% WER

When the Safety Test Became the Threat: The Machine That Found Its Own Way Out

OrcaRouter Releases OrcaCyber Zero 1.5 Cybersecurity Model With 1M Context

Sakana AI’s LLM Peer Review System Catches 73% of Core-Claim Errors

OpenAI Decisions API Hits Public Beta With 10x Faster Typed Answers

Microsoft AI Releases Microsoft-Decision-1: A Qwen3.5-9B Decision-Scoring Model

Nace AI Open-Sources Drex 1.5: A 9B Decision Model That Scores Options, Not Text

Alibaba Qwen Releases Qwen-Image-2.1-Turbo, an 8-Step 7B Image Model

Meet the Underdog Saluki 27B: A 2-bit Qwen3.8-27B That Beats the Original at Tool Calling

Google Research RRSI Guide: Mastering Self-Improving AI Agents

JetBrains 发布 Mellum2.1：面向编码智能体的 12B MoE 开源模型

Google Cloud Launches Gemini Agent, One Universal Agent for Enterprise Work

Architect 推出 Liquid Inference：为 LLM 推理引入实时竞价机制

Anthropic Releases Claude Haiku 5.5: A Small Model With 1M Context Priced at $0.10 per Million Input Tokens

What Happens When a Trusted Model Repo Changes? Unsloth Studio Re-Checks Before It Runs

NVIDIA 推出 PivotOPD：教多轮智能体从关键错误中恢复

Perplexity AI Releases pplx-embed-v2-late: A 0.6B Edge Model and a 9B Model Scoring 92.4% on MADQA

Liquid AI Releases Open-Weight d1-3B and d1-omni-600M: Multimodal Decision Models With Zero Output Tokens

Google DeepMind Releases EmbeddingGemma 2, a 740M Open Multimodal Embedding Model Built on Gemma 4