杨士海 × 江铸久 · Read19 2026全国围棋个人赛元老组第2轮 · 杨士海 vs 江铸久 · Move 93/93 · AI analysis, spoilers
Public professional games stream the board, win-rate trend, and move-by-move KataGo analysis. Watch live games as they happen, or reopen finished games after analysis is complete.
Live now
Read19 — Cloud KataGo and AI Go Review Online SGF Go Review | Read19 KataGo Go Analysis Sample | Read19 Play KataGo Online | Read19 Live Go Games With AI Analysis | Read19 Go Review Community | Read19 AI Go Style Profile | Read19 Ke Jie AI Go Style Profile | Read19 Zhan Ying AI Go Style Profile | Read19 Lian Xiao AI Go Style Profile | Read19 Read19 CLI — KataGo From Your Terminal KataGo API and KaTrain Setup | Read19 Read19 Service Status About Read19 Contact Read19 Privacy Policy | Read19 Terms of Service | Read19 Refund Policy | Read19 Read19 Support and Contact Turn an AI answer into a decision you can use next game. | Read19 Guides How one game error exposed a chain of service regressions | Read19 Does a Go-talking language model understand—or only sound as if it does? | Read19 Why can superhuman Go AI lose to seemingly weak play? | Read19 Which AI joseki retired, and why preserve them? | Read19 How far is AI-optimal from human-learnable? | Read19 Can komi or rules flip the AI's best move? | Read19 How an AI joseki moves from experiment to human habit | Read19 Does a KataGo answer have a shelf life? | Read19 AI-era joseki feels strangest when it exchanges less, tenukies early, and returns later | Read19 From New Fuseki to Chinese and Kobayashi systems, joseki became a component | Read19 Why the Shusaku diagonal held an entire order together before komi | Read19 What 5d, 9d, and professional policies see in the same Go position | Read19 Does giving KataGo thousands more visits really make the answer reliable | Read19 Twelve classic games show that the hardest move is not always the AI best move | Read19 The most interesting Go AI frontier is how it exposes good moves humans cannot see | Read19 How professionals review Go games with AI | Read19 Beyond the best move: Go AI should measure what a person can actually find | Read19 A decade of SOTA in Go AI from AlphaGo to transformer KataGo | Read19 How to choose a Go app: compare jobs, not feature counts | Read19 Choosing Fast, Balanced, Deep, and Human SL | Read19 Connect cloud KataGo to GUIs and scripts with Read19 CLI | Read19 GTP and JSONL are two KataGo languages on stdout | Read19 WSS is a channel, not new playing strength | Read19 KataGo is more than a model: a five-layer map | Read19 How to review a Go game with AI | Read19 From SGF source to a KataGo variation tree | Read19 How to connect KaTrain to Read19 cloud KataGo | Read19 AlphaGo turned joseki from answers into choices | Read19 Jowa's blood-vomiting game: three ghost moves under institutional pressure | Read19 The ear-reddening move faces four ways while AI calls it a loss | Read19 Go Seigen turns 3-3, star, and tengen into a manifesto | Read19 Shusai's sealed move 121 closes a retirement game and an era | Read19 Why the Atomic Bomb Game continued after the board was thrown | Read19 Cho Chikun from 0-3 to 4-3: a comeback made of four last chances | Read19 Nie Weiping versus Fujisawa: one board after a 7-7 team tie | Read19 Cho Hunhyun wins the first Ing Cup under eight-point komi | Read19 Rui Naiwei ties two dragons after defeating Cho Hunhyun | Read19 Lee Sedol runs a ladder he knows is broken: a 31-move whole-board ledger | Read19 AlphaGo's move 37 made everyone ask why | Read19 Lee Sedol's move 78 did not instantly reverse the game. It did something harder. | Read19