📄 文章列表
24 篇文章
cn_ai
Chrome 想终结“去邮箱复制验证码”，但隐私真的更好了吗？
cn_ai
百度开源文心5.5 Flash Base，多基准测试刷新同尺寸模型 SOTA
AlphaSignal
Unsloth Shrinks Qwen3.8-Flash-Next to 74.5 GB for Local Coding Agents
cn_ai
构建面向 Agentic 推理的下一代 LLM 推理引擎
VentureBeat
Qwen3.8-27B runs frontier-class coding agents and reasoning locally, no cloud API required
AlphaSignal
Unsloth's Qwen3.6 NVFP4 Hits 17,561 Tokens per Second on a Single GPU | AlphaSignal
AlphaSignal
Tencent's Hy3 Beats Flagship AI Models Without a Trillion Parameters | AlphaSignal
VentureBeat
DeepSeek open sources DSpark, a new framework to speed up LLM inference by up to 85%
AlphaSignal
Nvidia’s Open Salvo, OpenAI’s Amazon Deal, Grok Cuts Video Prices, Recursive Language Models
AlphaSignal
Prime Intellect Lets Teams Fine-Tune NVIDIA's Nemotron 3 Ultra Without a Single GPU | AlphaSignal
AlphaSignal
NVIDIA Shrinks GLM-5.2 Memory by 1.8x With NVFP4 Without Losing Accuracy | AlphaSignal
AlphaSignal
NVIDIA's Nemotron 3 Ultra Hits 6x Faster Inference With 550B Open Model | AlphaSignal
AlphaSignal
NVIDIA's Nemotron 3 Ultra Becomes the Most Capable Open US Model at 550B | AlphaSignal
AlphaSignal
Google Boosts Gemini Nano on Pixel 50% Faster Without Retraining | AlphaSignal
AlphaSignal
Z Lab's DFlash Hits 4.3x Throughput by Drafting Tokens in Parallel | AlphaSignal
AlphaSignal
SGLang Breaks 12,000 Tokens per Second on DeepSeek V4 Pro | AlphaSignal
AlphaSignal
Prime Intellect Lets Teams Fine-Tune NVIDIA's Nemotron 3 Ultra Without a Single GPU | AlphaSignal
AlphaSignal
NVIDIA's Nemotron 3 Ultra Hits 6x Faster Inference With 550B Open Model | AlphaSignal
AlphaSignal
NVIDIA's Nemotron 3 Ultra Becomes the Most Capable Open US Model at 550B | AlphaSignal
TDS
Training Objective That Makes LLM Inference 3X Faster
1
2