Commit Verlauf

Autor SHA1 Nachricht Datum
  Henry Linjamäki 8acdacb3ea opencl: use OpenCL C standard supported by the device (#12221) vor 11 Monaten
  John Bean 89b2b56e86 readme: added Sidekick to available UIs (#12311) vor 11 Monaten
  Georgi Gerganov e128a1bf5b tests : fix test-quantize-fns to init the CPU backend (#12306) vor 11 Monaten
  marcoStocchi 6ef79a67ca common : refactor '-o' option (#12278) vor 11 Monaten
  Olivier Chafik 4e39a3c332 `server`: extract <think> tags from qwq outputs (#12297) vor 11 Monaten
  Olivier Chafik be421fc429 `tool-call`: ensure there's always a non-empty tool call id (#12292) vor 11 Monaten
  Olivier Chafik 87c2630546 allow missing content in message if tool_calls provided (#12293) vor 11 Monaten
  Olivier Chafik 2b3a25c212 `sampler`: fixes trigger tokens + lazy grammars (fix typo cast from token to string) (#12291) vor 11 Monaten
  tc-mb 8352cdc87b llava : fix bug in minicpm-v code (#11513) vor 11 Monaten
  Georgi Gerganov 1e2f78a004 server : add speculative decoding presets for FIM (#12287) vor 11 Monaten
  Georgi Gerganov 0fd7ca7a21 authors : update (#12271) vor 11 Monaten
  Jason C.H 6fefc05a7a ggml-backend : make path_str compatible with C++20 (#12269) vor 11 Monaten
  Georgi Gerganov 7ab364390f server : infill gen ends on new line (#12254) vor 11 Monaten
  Daniel Bevenius 7c7f3b7f43 ggml : skip intermediate .air file when compiling .metallib (#12247) vor 11 Monaten
  Georgi Gerganov 102ac1891d sync : ggml vor 11 Monaten
  vmobilis d6ae2fa061 ggml : ggml_compute_forward_concat() for arbitrary tensor type (ggml/1118) vor 11 Monaten
  Rémy O 68d0027f3d ggml-cpu: faster AVX2 variant for IQ1_M (#12216) vor 11 Monaten
  Georgi Gerganov ea002810a2 ci : fix save-load test invocations (#12245) vor 11 Monaten
  Sigbjørn Skjæret 8fad3c7a7c server : Log original chat template parsing error (#12233) vor 11 Monaten
  Olivier Chafik 7cf64f6bee sync: minja - support QwQ-32B (#12235) vor 11 Monaten
  BB-fat 5e2d57b2b2 metal : simplify kernel arguments using a struct (#3229) (#12194) vor 11 Monaten
  David Huang f1648e91cf HIP: fix rocWMMA build flags under Windows (#12230) vor 11 Monaten
  Daniel Bevenius d6c95b0740 metal : fix default.metallib build (#12224) vor 11 Monaten
  lhez d76a86d967 opencl: Noncontiguous `norm`, `rms_norm`, disable `fp16` for some ops (#12217) vor 11 Monaten
  xiaofei 776f9e59cc cmake : fix undefined reference errors for std::filesystem in ggml (#12092) (#12094) vor 11 Monaten
  Lucas Moura Belo 3d652bfddf readme : update bindings (#12229) vor 11 Monaten
  Johannes Gäßler 5220a16d18 CUDA: fix FA logic for PTX 7.0 and CC >= 7.5 (#12222) vor 11 Monaten
  David Huang 3ffbbd5ce1 HIP: rocWMMA documentation and enabling in workflow builds (#12179) vor 11 Monaten
  Olivier Chafik 42994048a3 update function-calling.md w/ template override for functionary-small-v3.2 (#12214) vor 11 Monaten
  Aaron Teo e9b2f84f14 llava: add big-endian conversion for image encoder (#12218) vor 11 Monaten