1
0

Коммит түүх

Эзэн SHA1 Мессеж Огноо
  slaren da799b4189 vulkan : reuse parent extra for views (#7806) 1 жил өмнө
  Christian Zhou-Zheng c00fad71e5 gguf-split : change binary multi-byte units to decimal (#7803) 1 жил өмнө
  intelmatt 27615f5ab2 cmake : fix BUILD_SHARED_LIBS=ON build (#7784) 1 жил өмнө
  Johannes Gäßler 7027b27d76 server: update cache_prompt documentation [no ci] (#7745) 1 жил өмнө
  woodx a5cabd7649 server : do not get prompt in infill mode (#7286) 1 жил өмнө
  pengxin99 d5c938cd77 [SYCL] fix softmax r2r result wrong issue (#7811) 1 жил өмнө
  slaren c9ee7118d5 check for nans in imatrix and quantize (#7807) 1 жил өмнө
  Georgi Gerganov ee459f40f6 server : fix --threads-http arg (#7801) 1 жил өмнө
  Georgi Gerganov f83351f9a6 imatrix : migrate to gpt_params (#7771) 1 жил өмнө
  Clint Herron ad675e1c67 Added support for . (any character) token in grammar engine. (#6467) 1 жил өмнө
  Mattheus Chediak a143c04375 README minor fixes (#7798) [no ci] 1 жил өмнө
  Olivier Chafik 55b2d0849d grammars: x{min,max} repetition operator (#6640) 1 жил өмнө
  Joan Fontanals f5d7b268ec llama : add jina v2 base code (#7596) 1 жил өмнө
  slaren 2d08b7fbb4 docker : build only main and server in their images (#7782) 1 жил өмнө
  slaren d67caea0d6 docker : add openmp lib (#7780) 1 жил өмнө
  Galunid 7672adeec7 Fix encoding in python scripts (#7733) 1 жил өмнө
  Johannes Gäßler 7d1a378b8f CUDA: refactor mmq, dmmv, mmvq (#7716) 1 жил өмнө
  Georgi Gerganov 2b3389677a ggml : refactor rope norm/neox (#7634) 1 жил өмнө
  arch-btw 9973e81c5c readme : remove -ins (#7759) 1 жил өмнө
  jaime-m-p c90dbe026b Fix per token atrributes bits (#7749) 1 жил өмнө
  agray3 b90dc566c1 Allow number of nodes in CUDA graph to change (#7738) 1 жил өмнө
  Georgi Gerganov 1442677f92 common : refactor cli arg parsing (#7675) 1 жил өмнө
  Georgi Gerganov 554c247caf ggml : remove OpenCL (#7735) 1 жил өмнө
  Georgi Gerganov 0cd6bd3483 llama : remove beam search (#7736) 1 жил өмнө
  Georgi Gerganov 5ca0944a15 readme : remove obsolete Zig instructions (#7471) 1 жил өмнө
  slaren adc9ff3841 llama-bench : allow using a different printer for stderr with -oe (#7722) 1 жил өмнө
  Daniele 987d743d6b Improve hipBLAS support in CMake (#7696) 1 жил өмнө
  zhouwg b226c1227b refine .gitignore (#7688) 1 жил өмнө
  jaime-m-p 3b38d48609 Per token attributes (#7685) 1 жил өмнө
  Georgi Gerganov 6d1616944d ggml : prevent builds with -ffinite-math-only (#7726) 1 жил өмнө