Commit History

Autor SHA1 Mensaxe Data
  Daniel Hiltgen 3518aaef33 Merge pull request #4218 from dhiltgen/auto_parallel hai 10 meses
  Blake Mizerany cb42e607c5 llm: speed up gguf decoding by a lot (#5246) hai 10 meses
  Daniel Hiltgen 17b7186cd7 Enable concurrency by default hai 1 ano
  Daniel Hiltgen 5bf5aeec01 Refine mmap default logic on linux hai 10 meses
  Daniel Hiltgen 96624aa412 Merge pull request #5072 from dhiltgen/windows_path hai 10 meses
  Daniel Hiltgen 7784ca33ce Tighten up memory prediction logging hai 10 meses
  Daniel Hiltgen 171796791f Adjust mmap logic for cuda windows for faster model load hai 10 meses
  Daniel Hiltgen b2799f111b Move libraries out of users path hai 10 meses
  Daniel Hiltgen da3bf23354 Workaround gfx900 SDMA bugs hai 11 meses
  Daniel Hiltgen 6f351bf586 review comments and coverage hai 11 meses
  Daniel Hiltgen fc37c192ae Refine CPU load behavior with system memory visibility hai 11 meses
  Daniel Hiltgen 6fd04ca922 Improve multi-gpu handling at the limit hai 11 meses
  Craig Hughes b84aea1685 Critical fix from llama.cpp JSON grammar to forbid un-escaped escape characters inside strings, which breaks parsing. (#3782) hai 10 meses
  Michael Yang e40145a39d lint hai 11 meses
  Michael Yang c895a7d13f some gocritic hai 11 meses
  Michael Yang 829ff87bd1 revert tokenize ffi (#4761) hai 11 meses
  Jeffrey Morgan a50a87a7b8 partial offloading: allow flash attention and disable mmap (#4734) hai 11 meses
  Michael Yang 26a00a0410 use ffi for tokenizing/detokenizing hai 11 meses
  Daniel Hiltgen 92c81e8117 Give the final model loading more time hai 11 meses
  Lei Jitang 7487229c34 llm/server.go: Fix 2 minor typos (#4661) hai 11 meses
  Daniel Hiltgen 0165ba1651 Merge pull request #4638 from dhiltgen/better_error hai 11 meses
  Daniel Hiltgen c4209d6d21 Report better warning on client closed abort of load hai 11 meses
  Patrick Devine 4cc3be3035 Move envconfig and consolidate env vars (#4608) hai 11 meses
  Daniel Hiltgen b37b496a12 Wire up load progress hai 11 meses
  Jeffrey Morgan 38255d2af1 Use flash attention flag for now (#4580) hai 11 meses
  Sam e15307fdf4 feat: add support for flash_attn (#4120) hai 11 meses
  Patrick Devine d1692fd3e0 fix the cpu estimatedTotal memory + get the expiry time for loading models (#4461) hai 11 meses
  Daniel Hiltgen 853ae490e1 Sanitize the env var debug log hai 11 meses
  Patrick Devine 6845988807 Ollama `ps` command for showing currently loaded models (#4327) hai 11 meses
  jmorganca 92ca2cca95 Revert "only forward some env vars" hai 11 meses