Python check requirements.txt

Workflow runs for Python check requirements.txt

• hex-q1/q2: add support for flat and tiled GET_ROWS for Q1_0 and Q2_0

  completed successfully:
  Python check requirements.txt#290:Commitf3336cdpushedbymax-krasnyansky
  tb/hexagon-q2_0-support

  Oct 18, 20262m 34s
  More actions for workflow run 290

• ggml-cuda: HIP: use __builtin_amdgcn_perm for q1_0 unpacking (#29927)

  completed successfully:
  Python check requirements.txt#289:Commit9d1878apushedbymax-krasnyansky
  master

  Oct 17, 20262m 36s
  More actions for workflow run 289

• hexagon: add support for q2_0 quant

  completed successfully:
  Python check requirements.txt#288:Commitaeced3fpushedbytboinovski1
  tb/hexagon-q2_0-support

  Oct 17, 20262m 47s
  More actions for workflow run 288

• opencl: use q4_K in function names, move decls around

  completed successfully:
  Python check requirements.txt#287:Commita5c4f57pushedbylhez
  lh/cl-split

  Oct 16, 20262m 49s
  More actions for workflow run 287

• hexagon: enable Q1_0 weights in the 2D HMX MUL_MAT path

  completed successfully:
  Python check requirements.txt#286:Commit9ecbbe3pushedbytboinovski1
  tb/hexagon-q1_0-support

  Oct 15, 20262m 54s
  More actions for workflow run 286

• convert : fix Granite Switch block_count for MultiSwitch checkpoints …

  completed successfully:
  Python check requirements.txt#285:Commit59d5fadpushedbymax-krasnyansky
  master

  Oct 15, 20262m 44s
  More actions for workflow run 285

• tests: disable huge MM tests

  completed successfully:
  Python check requirements.txt#284:Commit486059cpushedbymax-krasnyansky
  hexagon-repack-rework

  Oct 15, 20262m 57s
  More actions for workflow run 284

• hex-get-rows: one more pass at recovering get-rows perf regression

  completed successfully:
  Python check requirements.txt#283:Commit2484b4apushedbymax-krasnyansky
  hexagon-repack-rework

  Oct 15, 20262m 58s
  More actions for workflow run 283

• opencl: use q4_K in function names, move decls around

  completed successfully:
  Python check requirements.txt#282:Commitd9a146dpushedbylhez
  lh/cl-split

  Oct 14, 20263m 0s
  More actions for workflow run 282

• llama : fix on-device state copy sizes for quantized KV caches

  completed successfully:
  Python check requirements.txt#281:Commit0e3dbf0pushedbywanghqc
  hq/llama-state-device-quant-size-up

  Oct 13, 20262m 45s
  More actions for workflow run 281

• opencl: dp4a MoE GEMMs, MoE on the Adreno 850, f16/bf16 experts

  completed successfully:
  Python check requirements.txt#280:Commit037404dpushedbywanghqc
  hq/opencl-consol-4-moe

  Oct 13, 20262m 41s
  More actions for workflow run 280

• opencl: Adreno correctness fixes

  completed successfully:
  Python check requirements.txt#279:Commit424f2e5pushedbywanghqc
  hq/opencl-consol-1-fixes

  Oct 13, 20262m 58s
  More actions for workflow run 279

• opencl: decode and prefill flash attention on Adreno

  completed successfully:
  Python check requirements.txt#278:Commita5d75e6pushedbywanghqc
  hq/opencl-consol-5-fa

  Oct 13, 20262m 58s
  More actions for workflow run 278

• opencl: write the gated_delta_net state snapshots straight into the r…

  completed successfully:
  Python check requirements.txt#277:Commit5f2b914pushedbywanghqc
  hq/opencl-gdn-cache-fusion-up

  Oct 13, 20262m 54s
  More actions for workflow run 277

• opencl: int8 fused KQ: stage Q per head-size slice and allow head siz…

  completed successfully:
  Python check requirements.txt#276:Commitadb1464pushedbywanghqc
  hq/opencl-int8-kq-dk512-up

  Oct 13, 20262m 43s
  More actions for workflow run 276

• opencl: low-bit and I-quant matrix multiply

  completed successfully:
  Python check requirements.txt#275:Commit6760f44pushedbywanghqc
  hq/opencl-consol-3-lowbit

  Oct 13, 20262m 50s
  More actions for workflow run 275

• opencl: fuse the residual add into the following rms_norm * w

  completed successfully:
  Python check requirements.txt#274:Commitc78d13apushedbywanghqc
  hq/opencl-consol-7-fusion

  Oct 13, 20262m 41s
  More actions for workflow run 274

• opencl: decode gemma-4 global attention through the FA decomposition …

  completed successfully:
  Python check requirements.txt#273:Commitc30d7dbpushedbywanghqc
  hq/opencl-gemma4-dk512-decode-decompose-up

  Oct 13, 20262m 54s
  More actions for workflow run 273

• opencl: read the gated_delta_net input state straight from the recurr…

  completed successfully:
  Python check requirements.txt#272:Commitc87b641pushedbywanghqc
  hq/opencl-gdn-state-fusion-up

  Oct 13, 20262m 55s
  More actions for workflow run 272

• opencl: build one program per kernel where the compiler cannot hold s…

  completed successfully:
  Python check requirements.txt#271:Commite737867pushedbywanghqc
  hq/opencl-consol-8-a6x

  Oct 13, 20262m 57s
  More actions for workflow run 271

• opencl: chunkwise gated delta net prefill and mamba2 ssm_scan row fold

  completed successfully:
  Python check requirements.txt#270:Commitd29262fpushedbywanghqc
  hq/opencl-consol-6-ssm

  Oct 13, 20263m 8s
  More actions for workflow run 270

• opencl: int8 GEMM staging, small-batch verify kernels and weight plac…

  completed successfully:
  Python check requirements.txt#269:Commitf652bccpushedbywanghqc
  hq/opencl-consol-2-dp4a

  Oct 13, 20262m 42s
  More actions for workflow run 269

• opencl: use q4_K in function names, move decls around

  completed successfully:
  Python check requirements.txt#268:Commitb89151fpushedbylhez
  lh/cl-split

  Oct 9, 20262m 49s
  More actions for workflow run 268

• sycl : Q5_K reorder-layout MMVQ and fused GLU (#29375)

  completed successfully:
  Python check requirements.txt#267:Commit3d65c90pushedbylhez
  master

  Oct 9, 20262m 52s
  More actions for workflow run 267

• llama: re-create the KV cache when flash attention resolves to disabled

  completed successfully:
  Python check requirements.txt#266:Commitea909ecpushedbywanghqc
  hq/llama-fa-auto-recreate-kv-cache

  Oct 8, 20262m 51s
  More actions for workflow run 266