📄 文章列表
25 篇文章
AlphaSignal
Latent Sparse Attention Speeds Up 10M-Token AI Contexts by 9x
AlphaSignal
Meta Releases Muse Ember, the Cheapest Agent Model at $0.14 a Task
AlphaSignal
Gemini 4 Carbon Tops Scientific Vision Benchmark With 90x Cost Gap
VentureBeat
Thinking Machines launches Inkling 2 Small with near-frontier performance on enterprise benchmarks
AlphaSignal
Artificial Analysis Finds 1 in 4 AI Endpoints Quietly Underperform Their Advertised Models
cn_ai
Token 暴涨 5000 倍之后,企业 Agent 落地还卡在哪里?
cn_ai
混元 Hy4 实测:腾讯的产品基因,模型继承了几分?
AlphaSignal
Artificial Analysis Ditches Tokens per Second to Rank AI Speed
AlphaSignal
Engram's RL Training Teaches Agents to Read Their Own Notes 20% Better
cn_ai
同一个模型从 30 分到 95 分:拉开差距的不是模型,是Harness
AlphaSignal
Humans& Ships Weight Teleportation to Sync 445 GB AI Models in 220 ms
AlphaSignal
Mistral AI Drops Mistral Large 4, a 1T Open-Weight Coding Beast
cn_ai
Claude紧急叫停!暗改Harness禁止逆向游戏
AlphaSignal
humans& Trains Ampersand Tutor by Teaching a Simulated Student
cn_ai
天数智芯公布天垓300实测数据:部分推理场景达到H200水平,端侧对标Thor T5000
VentureBeat
Mistral Large 4 'Le Chonk' arrives as the top open model outside China with improved coding performance and 63% lower co
VentureBeat
A coding agent’s harness can triple its cost without making it more accurate, study finds
VentureBeat
MiniMax's new M3.1 Flash beats DeepSeek and Opus 5 on a key business benchmark at lower cost — but with a high hallucina
VentureBeat
Meta unveils Muse Spark Open, the new top U.S. open weights model — and it runs on two paired Mac desktops
VentureBeat
NVIDIA's new, open Nemotron 3.5 Super VL with post-training beats Claude Opus 5, GPT-6.1 Sol on specialized task benchma
1
2