Commit Verlauf

Autor SHA1 Nachricht Datum
  xiaobing318 1a99c2d948 cmake : fix target include directories (#15450) vor 5 Monaten
  Daniel Bevenius 37f10f955f make : remove make in favor of CMake (#15449) vor 5 Monaten
  Georgi Gerganov 2f37014073 lookahead : add sample command to readme (#15447) vor 5 Monaten
  R0CKSTAR a094f38143 musa: fix build warnings (#15258) vor 5 Monaten
  lhez fb22dd07a6 opencl: mark `argsort` unsupported if cols exceed workgroup limit (#15375) vor 5 Monaten
  Georgi Gerganov 9ef6b0b835 model : add gpt-oss type strings (#15424) vor 5 Monaten
  Gian-Carlo Pascutto 1e19f5d462 common : Add top-nsigma sampler to help globally (#15428) vor 5 Monaten
  Georgi Gerganov d2fcd91cf9 server : disable context shift by default (#15416) vor 5 Monaten
  SHUAI YANG a6d3cfe7fa CANN: optimize rope operator (#15335) vor 5 Monaten
  R0CKSTAR 67f09a3a27 musa: handle __hgt2_mask, available starting from MUSA SDK rc4.3.0 (#15413) vor 5 Monaten
  Marvin Gießing 6424594c56 ggml-cpu: add mxfp4 VSX intrinsics for Power9+ (ppc64le) hardware (#15385) vor 5 Monaten
  Xuan-Son Nguyen e9288e8869 chat : clarify the meaning of reasoning_format (#15408) vor 5 Monaten
  Georgi Gerganov 9d262f4bad server : remove swa_full warning (#15399) vor 5 Monaten
  Georgi Gerganov f0d3c7405c batched-bench : use rand tokens (#15398) vor 5 Monaten
  Xuan-Son Nguyen f08c4c0d8d mtmd : clean up clip_n_output_tokens (#15391) vor 5 Monaten
  Georgi Gerganov 6d7f1117e3 codeowners : remove mmv.* vor 5 Monaten
  Georgi Gerganov 60212f1ead sync : ggml vor 5 Monaten
  Georgi Gerganov f0c541d315 scripts : update sync scripts vor 5 Monaten
  Sigbjørn Skjæret baa9255a45 llama : merge conts and reshapes and remove unnecessary cont (#15380) vor 5 Monaten
  Georgi Gerganov 3007baf201 readme : update hot topics (#15397) vor 5 Monaten
  davidef d1d8241600 server : fix incoming tasks not process in order (#15395) vor 5 Monaten
  Dobri Danchev 618575c582 Fix broken build: require updated pip to support --break-system-packages (#15357) vor 5 Monaten
  compilade f44f793172 ggml-quants : fix make_qp_quants NANs and IQ1 assertion errors (#15379) vor 5 Monaten
  Jeff Bolz ae532eac2c vulkan: disable spirv-opt for bfloat16 shaders (#15352) vor 5 Monaten
  Oleksandr Kuvshynov e5155e6986 server : export max observed n_past value (#15361) vor 5 Monaten
  Jeff Bolz 21c17b5bef vulkan: Use larger workgroups for mul_mat_vec when M is small (#15355) vor 5 Monaten
  Dong Won Kim 19f4decae0 vulkan: support sqrt (#15370) vor 5 Monaten
  Sigbjørn Skjæret 4d196981d4 convert : force patch_embd weights to F16 or F32 to avoid broken GGUFs (#15367) vor 5 Monaten
  Sigbjørn Skjæret b143fbc87a ci : fix hang in windows-hip build/release (#15365) vor 5 Monaten
  Jeff Bolz de5627910d vulkan: Optimize argsort (#15354) vor 5 Monaten