Commit History

Author SHA1 Message Date
  Daniel Hiltgen a3c20e3f18 Refine error reporting for subprocess crash 9 months ago
  Daniel Hiltgen 283948c83b Adjust windows ROCm discovery 9 months ago
  royjhan b9f5e16c80 Introduce `/api/embed` endpoint supporting batch embedding (#5127) 9 months ago
  Jeffrey Morgan ef98803d63 llm: looser checks for minimum memory (#5677) 9 months ago
  Jeffrey Morgan c4cf8ad559 llm: avoid loading model if system memory is too small (#5637) 9 months ago
  Jeffrey Morgan 791650ddef sched: only error when over-allocating system memory (#5626) 9 months ago
  Daniel Hiltgen 22c81f62ec Remove duplicate merge glitch 9 months ago
  Michael Yang 9bbddc37a7 Merge pull request #5126 from ollama/mxyng/messages 9 months ago
  Jeffrey Morgan 53da2c6965 llm: remove ambiguous comment when putting upper limit on predictions to avoid infinite generation (#5535) 9 months ago
  Michael Yang ac7a842e55 fix model reloading 10 months ago
  Daniel Hiltgen ccd7785859 Merge pull request #5243 from dhiltgen/modelfile_use_mmap 10 months ago
  Daniel Hiltgen 0e982bc1f4 Fix corner cases on tmp cleaner on mac 10 months ago
  Josh Yan 33a65e3ba3 error 10 months ago
  Daniel Hiltgen 97c9e11768 Switch use_mmap to a pointer type 10 months ago
  Daniel Hiltgen 3518aaef33 Merge pull request #4218 from dhiltgen/auto_parallel 10 months ago
  Blake Mizerany cb42e607c5 llm: speed up gguf decoding by a lot (#5246) 10 months ago
  Daniel Hiltgen 17b7186cd7 Enable concurrency by default 1 year ago
  Daniel Hiltgen 5bf5aeec01 Refine mmap default logic on linux 10 months ago
  Daniel Hiltgen 96624aa412 Merge pull request #5072 from dhiltgen/windows_path 10 months ago
  Daniel Hiltgen 7784ca33ce Tighten up memory prediction logging 10 months ago
  Daniel Hiltgen 171796791f Adjust mmap logic for cuda windows for faster model load 10 months ago
  Daniel Hiltgen b2799f111b Move libraries out of users path 10 months ago
  Daniel Hiltgen da3bf23354 Workaround gfx900 SDMA bugs 11 months ago
  Daniel Hiltgen 6f351bf586 review comments and coverage 11 months ago
  Daniel Hiltgen fc37c192ae Refine CPU load behavior with system memory visibility 11 months ago
  Daniel Hiltgen 6fd04ca922 Improve multi-gpu handling at the limit 11 months ago
  Craig Hughes b84aea1685 Critical fix from llama.cpp JSON grammar to forbid un-escaped escape characters inside strings, which breaks parsing. (#3782) 10 months ago
  Michael Yang e40145a39d lint 11 months ago
  Michael Yang c895a7d13f some gocritic 11 months ago
  Michael Yang 829ff87bd1 revert tokenize ffi (#4761) 11 months ago