Commit History

Auteur SHA1 Bericht Datum
  Patrick Devine 4cc3be3035 Move envconfig and consolidate env vars (#4608) 11 maanden geleden
  Sang Park 4434d7f447 Correct typo in error message (#4535) 11 maanden geleden
  Daniel Hiltgen ec231a7923 Remove VRAM convergence check for windows 11 maanden geleden
  Patrick Devine 6845988807 Ollama `ps` command for showing currently loaded models (#4327) 11 maanden geleden
  Daniel Hiltgen 4142c3ef7c Always use the sorted list of GPUs 11 maanden geleden
  Jeffrey Morgan bb6fd02298 Don't clamp ctx size in `PredictServerFit` (#4317) 11 maanden geleden
  Daniel Hiltgen 354ad9254e Wait for GPU free memory reporting to converge 11 maanden geleden
  Jeffrey Morgan c9f98622b1 Skip scheduling cancelled requests, always reload unloaded runners (#4189) 1 jaar geleden
  Jeffrey Morgan dfa2f32ca0 unload in critical section (#4187) 1 jaar geleden
  Daniel Hiltgen f56aa20014 Centralize server config handling 1 jaar geleden
  Jeffrey Morgan 942c979232 allocate a large enough kv cache for all parallel requests (#4162) 1 jaar geleden
  Daniel Hiltgen 20f6c06569 Make maximum pending request configurable 1 jaar geleden
  Mark Ward 63c763685f log when the waiting for the process to stop to help debug when other tasks execute during this wait. 1 jaar geleden
  Mark Ward f4a73d57a4 fix runner expire during active use. Clearing the expire timer as it is used. Allowing the finish to assign an expire timer so that the runner will expire after no use. 1 jaar geleden
  Daniel Hiltgen d6e3b64582 Fix concurrency for CPU mode 1 jaar geleden
  Daniel Hiltgen 9b5a3c5991 Merge pull request #3914 from dhiltgen/mac_perf 1 jaar geleden
  Jeffrey Morgan 00b0699c75 Reload model if `num_gpu` changes (#3920) 1 jaar geleden
  Daniel Hiltgen b123be5b71 Adjust context size for parallelism 1 jaar geleden
  Bryce Reitano 36a6daccab Restructure loading conditional chain 1 jaar geleden
  Bryce Reitano 284e02bed0 Move ggml loading to when we attempt fitting 1 jaar geleden
  Daniel Hiltgen 34b9db5afc Request and model concurrency 1 jaar geleden