Commit History

Autor SHA1 Mensaxe Data
  Daniel Hiltgen 34b9db5afc Request and model concurrency hai 1 ano
  Michael Yang 26df674785 scale graph based on gpu count hai 1 ano
  Michael Yang 41a272de9f darwin: no partial offloading if required memory greater than system hai 1 ano
  Michael Yang 7e33a017c0 partial offloading hai 1 ano
  Daniel Hiltgen be330174dd Allow setting max vram for workarounds hai 1 ano
  peanut256 a189810df6 Determine max VRAM on macOS using `recommendedMaxWorkingSetSize` (#2354) hai 1 ano
  Daniel Hiltgen 7427fa1387 Fix up the CPU fallback selection hai 1 ano
  Daniel Hiltgen 39928a42e8 Always dynamically load the llm server library hai 1 ano
  Daniel Hiltgen d88c527be3 Build multiple CPU variants and pick the best hai 1 ano
  Jeffrey Morgan c336693f07 calculate overhead based number of gpu devices (#1875) hai 1 ano
  Jeffrey Morgan 08f1e18965 Offload layers to GPU based on new model size estimates (#1850) hai 1 ano
  Jeffrey Morgan c7ea8f237e set `num_gpu` to 1 only by default on darwin arm64 (#1771) hai 1 ano
  Daniel Hiltgen a2ad952440 Fix windows system memory lookup hai 1 ano
  Daniel Hiltgen d966b730ac Switch windows build to fully dynamic hai 1 ano
  Daniel Hiltgen 7555ea44f8 Revamp the dynamic library shim hai 1 ano
  Daniel Hiltgen 6558f94ed0 Fix darwin intel build hai 1 ano
  Daniel Hiltgen 35934b2e05 Adapted rocm support to cgo based llama.cpp hai 1 ano