GitLab

fix HEAD request

silence warm up log

HEAD request should respond like their GET counterparts except without a
response body.

* remove tmp directories created by previous servers

* clean up on server stop

* Update routes.go

* Update server/routes.go
Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com>

* create top-level temp ollama dir

* check file exists before creating

---------
Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com>
Co-authored-by: Michael Yang <mxyng@pm.me>

Use API to check if model exists and pull if necessary

this should be less error prone

fix OLLAMA_HOST parsing for ip6

pack in cuda libs

fix impossible condition

explicitly follow upload redirects

fix build

fix mkdir on windows

* subprocess improvements

- increase start-up timeout
- when runner fails to start fail rather than timing out
- try runners in order rather than choosing 1 runner
- embed metal runner in metal dir rather than gpu
- refactor logging and error messages

* Update llama.go

* Update llama.go

* simplify by using glob

fix error on upload chunk

implement ProgressWriter

set request.ContentLength

This informs the HTTP client the content length is known and disables
chunked Transfer-Encoding