Repository Issue Activity (beta)

#916 A question in the "match" module

#1238 [Discussion / Feedback] ONNX Backend: Benchmarks, Questions, and Issues

#432 No response for several hours when running contribute mode

#1248 ROCm backend on Windows fails to create MIOpen cache directory with non-ASCII username

#1207 Incompatible with TensorRT 11

#1249 Feature/API idea: temporarily prioritize a specific root move without clearing the current search tree

#1251 Transformer model performance on Apple Neural Engine

#1245 ROCm 7.2 + RDNA4: performance silently collapses (CK FMHA probe fails, ~2x slower)

#1236 In katago selfplay, are the numSearchThreads and the value shown in the benchmark the same?

#1247 About interleaved RoPE in transformer models

#1244 Question about the latest KataGo weight file

#1242 S

#1035 How to load xx.bin.gz on Andorid?

#1221 KataGo 1.17.0 `contribute` crash: uncaught `std::length_error` during OpenCL xGemmDirect tuning on AMD gfx1201

#1215 [News] Shin Jin-seo 2:1 KataGo with two handicaps, 10 years later AlphaGo

#1214 A Belated Correction for the 8x8 Opening Book

#333 Using symmetry

#1080 Integrating attention and feature vectors?

#1227 Persistent bug: score lead contradiction that cannot be corrected

#1220 [Feature Request] Transformer models are amazing on TRT, but OpenCL leaves 80%+ performance on the table – can we get a better general backend? (I can help test!)

#1224 User Test Report: b11c768 Transformer Model vs b28c512nbt on RTX 3080

#1233 Could you give a rough estimate of the Elo difference between the new 15b weights and the 11b at the same visits? Which one is expected to have the advantage at the same compute time, and by how many Elo points? Thanks in advance!

#1232 When will the b15c1024 model in the Transformer series be updated

#1182 Analysis Bug: Severe miscalculation in multiple strongest models on famous ancient game (Fan Xiping vs Huang Yougong)

#1161 An Open Letter to lightvector: New Year Wishes, Community Breakthroughs & My Hopes for KataGo in 2026