Cronologia Commit

Autore SHA1 Messaggio Data
  Daniel Hiltgen 283948c83b Adjust windows ROCm discovery 9 mesi fa
  royjhan b9f5e16c80 Introduce `/api/embed` endpoint supporting batch embedding (#5127) 9 mesi fa
  Jeffrey Morgan ef98803d63 llm: looser checks for minimum memory (#5677) 9 mesi fa
  Jeffrey Morgan c4cf8ad559 llm: avoid loading model if system memory is too small (#5637) 9 mesi fa
  Jeffrey Morgan 791650ddef sched: only error when over-allocating system memory (#5626) 9 mesi fa
  Daniel Hiltgen 22c81f62ec Remove duplicate merge glitch 9 mesi fa
  Michael Yang 9bbddc37a7 Merge pull request #5126 from ollama/mxyng/messages 9 mesi fa
  Jeffrey Morgan 53da2c6965 llm: remove ambiguous comment when putting upper limit on predictions to avoid infinite generation (#5535) 9 mesi fa
  Michael Yang ac7a842e55 fix model reloading 10 mesi fa
  Daniel Hiltgen ccd7785859 Merge pull request #5243 from dhiltgen/modelfile_use_mmap 10 mesi fa
  Daniel Hiltgen 0e982bc1f4 Fix corner cases on tmp cleaner on mac 10 mesi fa
  Josh Yan 33a65e3ba3 error 10 mesi fa
  Daniel Hiltgen 97c9e11768 Switch use_mmap to a pointer type 10 mesi fa
  Daniel Hiltgen 3518aaef33 Merge pull request #4218 from dhiltgen/auto_parallel 10 mesi fa
  Blake Mizerany cb42e607c5 llm: speed up gguf decoding by a lot (#5246) 10 mesi fa
  Daniel Hiltgen 17b7186cd7 Enable concurrency by default 1 anno fa
  Daniel Hiltgen 5bf5aeec01 Refine mmap default logic on linux 10 mesi fa
  Daniel Hiltgen 96624aa412 Merge pull request #5072 from dhiltgen/windows_path 10 mesi fa
  Daniel Hiltgen 7784ca33ce Tighten up memory prediction logging 10 mesi fa
  Daniel Hiltgen 171796791f Adjust mmap logic for cuda windows for faster model load 10 mesi fa
  Daniel Hiltgen b2799f111b Move libraries out of users path 10 mesi fa
  Daniel Hiltgen da3bf23354 Workaround gfx900 SDMA bugs 11 mesi fa
  Daniel Hiltgen 6f351bf586 review comments and coverage 11 mesi fa
  Daniel Hiltgen fc37c192ae Refine CPU load behavior with system memory visibility 11 mesi fa
  Daniel Hiltgen 6fd04ca922 Improve multi-gpu handling at the limit 11 mesi fa
  Craig Hughes b84aea1685 Critical fix from llama.cpp JSON grammar to forbid un-escaped escape characters inside strings, which breaks parsing. (#3782) 10 mesi fa
  Michael Yang e40145a39d lint 11 mesi fa
  Michael Yang c895a7d13f some gocritic 11 mesi fa
  Michael Yang 829ff87bd1 revert tokenize ffi (#4761) 11 mesi fa
  Jeffrey Morgan a50a87a7b8 partial offloading: allow flash attention and disable mmap (#4734) 11 mesi fa