资讯与观察· 模型发布
AI 行业动态、工程实践、产品方法和社区故事。
Tsinghua AI Open-Sources AKV, a Cache Engineering Server for LLM Agents on SGLang
Tsinghua researchers released AKV, an SGLang-based serving system for cache engineering. LLM agents drop and reposition KV cache instead of summarizing history, cutting average cost 27.7% across 4 agent benchmarks.
How Fanatics and OpenAI streamline the fan experience
Fanatics uses OpenAI to help fans complete their game-day outfits with Shop the Look. Testing across roughly 30,000 MLB products showed 10% higher engagement.
6 分钟 · OpenAI
Nanbeige Releases Nanbeige4.3-33B, a 33B Open Agentic Coding Model
Nanbeige LLM Lab released Nanbeige4.3-33B, a 33B dense Apache 2.0 model with 512K context. It reports 80.9 on Terminal-Bench 2.1 and will teach the upcoming Nanbeige4.3-3B.
14 分钟 · MarkTechPostLATEST STORIES
最新内容
Hugging Face Releases Carbon-X, a 14B Open DNA Language Model that Brings DNA and Natural Language Together for the First Time精选导读
Hugging Face released Carbon-X, an open 14.8B DNA language model built on Qwen3-14B. It reads raw DNA in chat, answers in English, and averages 66.7% across 5 DNA benchmarks under Apache 2.0.
Researchers from Stanford and UC Berkeley Introduce TokenGate: Token-Gated Attention Makes Long-Context Prefill Up to 11.3x Faster精选导读
Stanford and UC Berkeley researchers released TokenGate, a drop-in learned token pruning module for LLMs. It cuts long-context prefill time-to-first-token by 1.4x to 11.3x and peak memory by 1.5x to 3.4x without accurac…
Anthropic Launches OSS Scanner, a Free AI Vulnerability Scanner精选导读
Anthropic has launched OSS Scanner, a free, opt-in vulnerability scanner for critical open-source projects. The Anthropic OSS Scanner runs the company’s strongest models, including Claude Mythos, against enrolled codeba…
Microsoft Open-Sources Agensh: A Multi-Agent Harness for 1,024 Agents精选导读
Microsoft Research open-sourced Agensh, a multi-agent harness with no central orchestrator. Self-organizing workers share a Git repo, chat and a findings board, scaling from 1 to 1,024 agents on ProgramBench under a 6-h…
Cognition Open-Sources an Agent Plugin that Helps Secure Systems against Recent Cybersecurity Threats精选导读
Cognition released the Devin Vulnerability Fixes plugin, 3 MIT-licensed agent skills that inventory, patch and test PHP, Node.js and self-managed GitLab deployments hit by this week’s critical RCE flaws, then open 1 PR …
How to Run Underdog Saluki 27B: 2-Bit Qwen3.8-27B Tool-Calling Guide精选导读
Explore a complete coding guide comparing Underdog Saluki 27B with ISTA's GSQ-RCO quantization, examining GGUF headers, memory usage, and BFCL tool-calling performance in stock llama.cpp
Meet Anthropic’s sandbox-runtime: OS-Level Sandboxing for AI Agents, No Containers精选导读
Anthropic sandbox-runtime (srt) is the open-source, OS-level sandbox behind Claude Code. It restricts file and network access for agents, MCP servers and CLI tools with Seatbelt on macOS and bubblewrap on Linux, no cont…
Stateful Agentic Engineering Copilots with Letta: Core Memory, Shared State, Semantic Retrieval, Custom Tools, Streaming, and Portable Agent Workflows精选导读
Discover how to extend agentic memory beyond basic prompts by building a stateful engineering copilot with Letta. This comprehensive guide covers inspectable core memory, archival vector retrieval, multi-agent shared st…
NVIDIA Releases Nemotron 3.5 Bolt E4B 4GB: An Open Small Language Model for Local Voice Agents, Game Characters and Robots精选导读
NVIDIA released Nemotron 3.5 Bolt E4B 4GB, an open small language model post-trained from Gemma 4 E4B for local voice assistants, game characters and robots, targeting about 4GB of VRAM on RTX and Jetson.