GitLab

Fixed improper env references

Forgot a " on the write-host

Added OLLAMA_CUSTOM_CUDA_DEFS and OLLAMA_CUSTOM_ROCM_DEFS to customize GPU builds on Windows

Added OLLAMA_CUSTOM_CUDA_DEFS and OLLAMA_CUSTOM_ROCM_DEFS instead of OLLAMA_CUSTOM_GPU_DEFS

add stablelm graph calculation

fix padding to only return padding

types/model: support : in PartHost for host:port

account for all non-repeating layers

Also, add test for DisplayLongest.

Also, plumb fill param to ParseName in MustParseName

scale graph based on gpu count

* parse wide argv characters on windows

* cleanup

* move cleanup to end of `main`

darwin: no partial offloading if required memory greater than system

TODO: update padding() to _only_ returning the padding

This reverts commit 7d05a6ee.

This proved to be more painful than useful.

See: https://github.com/ollama/ollama/issues/3624

Added just one line

| Solar              | 10.7B      | 6.1GB | `ollama run solar`             |

Changed ollama.call() for ollama.invoke() as per deprecated documentation from langchain

* Added MindsDB information

Added more details to MindsDB so that Ollama users can know that they can connect their Ollama model with 200+ databases and apps

* updated text for mindsdb

* examples: go-multimodal

* examples: add go-pull-progress

* examples: add go-chat

* fix

Remove Modelfile parameters that are decided at runtime

* terminate subprocess if receiving `SIGINT` or `SIGTERM` signals while model is loading

* use `unload` in signal handler