História revízii

Autor SHA1 Správa Dátum
  Daniel Hiltgen 92c81e8117 Give the final model loading more time 11 mesiacov pred
  Lei Jitang 7487229c34 llm/server.go: Fix 2 minor typos (#4661) 11 mesiacov pred
  Daniel Hiltgen 0165ba1651 Merge pull request #4638 from dhiltgen/better_error 11 mesiacov pred
  Daniel Hiltgen c4209d6d21 Report better warning on client closed abort of load 11 mesiacov pred
  Patrick Devine 4cc3be3035 Move envconfig and consolidate env vars (#4608) 11 mesiacov pred
  Daniel Hiltgen b37b496a12 Wire up load progress 11 mesiacov pred
  Jeffrey Morgan 38255d2af1 Use flash attention flag for now (#4580) 11 mesiacov pred
  Sam e15307fdf4 feat: add support for flash_attn (#4120) 11 mesiacov pred
  Patrick Devine d1692fd3e0 fix the cpu estimatedTotal memory + get the expiry time for loading models (#4461) 11 mesiacov pred
  Daniel Hiltgen 853ae490e1 Sanitize the env var debug log 11 mesiacov pred
  Patrick Devine 6845988807 Ollama `ps` command for showing currently loaded models (#4327) 11 mesiacov pred
  jmorganca 92ca2cca95 Revert "only forward some env vars" 11 mesiacov pred
  Daniel Hiltgen c4014e73a2 Fall back to CPU runner with zero layers 11 mesiacov pred
  Jeffrey Morgan bb6fd02298 Don't clamp ctx size in `PredictServerFit` (#4317) 11 mesiacov pred
  Michael Yang cf442cd57e fix typo 11 mesiacov pred
  Michael Yang ce3b212d12 only forward some env vars 11 mesiacov pred
  Michael Yang 58876091f7 log clean up 11 mesiacov pred
  Daniel Hiltgen d0425f26cf Merge pull request #4294 from dhiltgen/harden_subprocess_reaping 11 mesiacov pred
  Bruce MacDonald cfa84b8470 add done_reason to the api (#4235) 11 mesiacov pred
  Daniel Hiltgen 84ac7ce139 Refine subprocess reaping 11 mesiacov pred
  Daniel Hiltgen 920a4b0794 Merge remote-tracking branch 'upstream/main' into pr3702 11 mesiacov pred
  Daniel Hiltgen ee49844d09 Merge pull request #4153 from dhiltgen/gpu_verbose_response 11 mesiacov pred
  Daniel Hiltgen bee2f4a3b0 Record GPU usage information 1 rok pred
  Daniel Hiltgen 72700279e2 Detect noexec and report a better error 1 rok pred
  Daniel Hiltgen 380378cc80 Use our libraries first 1 rok pred
  Jeffrey Morgan ed740a2504 Fix `no slots available` error with concurrent requests (#4160) 1 rok pred
  Jeffrey Morgan 1b0e6c9c0e Fix llava models not working after first request (#4164) 1 rok pred
  Daniel Hiltgen f56aa20014 Centralize server config handling 1 rok pred
  Mark Ward 321d57e1a0 Removing go routine calling .wait from load. 1 rok pred
  Mark Ward ba26c7aa00 it will always return an error due to Kill() discarding Wait() errors 1 rok pred